Pouches, pockets and sacs in the heads, necks and chests of mammals, part II: elephants have a pouch in the throat... or do they?

ResearchBlogging.org

Welcome to the second part of the series on the various pouches, sacs and pockets present in the heads, necks and chests of mammals. Last time we looked at the laryngeal sacs of primates (and, should you encounter unfamiliar anatomical terms in the following text, be sure to check out that first article for an anatomical primer). Comparatively speaking, the structures present in primates are well known... or, at least, their existence is comparatively well known. Less well known is the suggestion that elephants possess a sort of pouch in the throat. Or... do they?

i-f9293d651f83deafdd68ffd820e0f1c7-Watson-1873-Elephas-pharyngeal-pouch-Oct-2010.jpg

Many people have reported elephants (both Asian and African) to reach into the mouth with the trunk and withdraw water. The elephants then spray this water over themselves (particularly their ears) to keep cool. The quantity of water produced in these instances is apparently quite large: J. Emerson Tennent, writing in his 1861 Sketches of the Natural History of Ceylon, referred to "gallons of water" being withdrawn. Richard Owen wrote about this phenomenon too, and most authors followed him in assuming or asserting that this water was either stored in a special compartment in the stomach, or was somehow regurgitated.

However, Watson (1873) described a large pharyngeal pouch, located both beneath and behind the base of the tongue. Its posterior border was, Watson described, formed by the thyroid cartilage of the larynx [Watson's anatomical drawing is show above: the pouch is the oval structure located in the middle: its opening is the large recess immediately above. The base of the tongue is shown at far left]. Watson suggested that this was where that stored water came from. His observations were later elaborated upon by Jeheskel Shoshani and colleagues, who described the pharyngeal pouch and its possible function in several articles (Shoshani 1998, Shoshani et al. 1997, Shoshani & Marchant 2001). They suggested that it could be used to carry as much as 1 gallon (3.8 kg) of water. I actually just went and looked at a gallon of liquid for a short while, just to remind myself how much water this actually is. It's a lot. The excellent diagram below shows the location and inferred function of the elephant's pouch; the diagram is by Gary H. Marchant and originally appeared in Shoshani (1998).

i-076cbcef5c704a7194d6c12ce6362446-elephant-pharyngeal-pouch-Shoshani.jpg

Shoshani et al. (1997) proposed that the pouch evolved in concert with the lowered larynx as a resonating chamber, and that its later use as an occasional site for water storage was an exaptation (an exaptation - or exapted structure - is a structure that evolved under selection for one role, and layer became co-opted for another. For previous discussions of exapted structures on Tet Zoo, go here on the gliding flaps of geckos, here on the jaw muscles of caecilians, and here on saurischian dinosaur tracheae and sterna).

i-59cc05360d08f72d4c2029f3c7c059dc-Elephas-hyoids-Shoshani-&-Marchant-Oct-2010.jpg

The pouch is partially supported by the hyoid skeleton and its associated musculature. And elephant hyoids are particularly unusual: the thyrohyals are delicate and strongly twisted along their length, and the stylohyals are really slender, with really long posterior and inferior rami [hyoids of Elephas shown here, from Shoshani & Marchant (2001)]. This distinctive anatomy means that certain inferences can be made about the soft tissue anatomy of the throat in extinct elephants (and other proboscideans) if their hyoids are well preserved. Shoshani et al. (1997) and Shoshani & Tassy (2005) concluded that a pouch like that of living elephants was present in all members of Elephantida: the proboscidean clade that includes gomphotheres and elephantids, but not mastodonts or more stem-ward proboscideans (here's that (much simplified) proboscidean cladogram again [previously used here]. It features artwork by Maurice Wilson, ZdenÄk Burian and Mauricio Antón].

i-6e6effe1cc58daa8dc4fd0c84194df8c-proboscideans-490-px-Oct-2010.jpg

One interesting question: if the pouch originally functioned in vocalising (as a resonating chamber), and if it can be used to hold water, how might the vocalising function and the water-holding function interact? So far as I know, no one has studied this in detail. However, Garstang (2004) - in an article on long-distance communication in elephants (via infrasound) - noted that the presence of water in the pharyngeal pouch might make a difference to the acoustic properties of elephant noises.

Ok, so far so good. It would appear that the existence of the elephant pharyngeal pouch is well established and fully verified. However...

Those of you who recall my review article on the elephant episode of Inside Nature's Giants series 1 may recall the mention there of the pharyngeal pouch, or, rather, the mention of the lack of mention. Joy Reidenberg noted in the comments that she failed to find the pouch during dissection, despite a deliberate search and prior knowledge of the Shoshani et al. research. She had to conclude that the so-called pouch was not a 'special' feature of elephants at all, but rather that it represented the space normally present in this position (the epiglottic valecula: the part of the mouth where chewed food is formed into a bolus for swallowing). So far as I know, Joy's observation is novel and has yet to be reported in the literature. In other words, you heard it here first [the impressive complexity of elephant throat musculature is well depicted in this diagram by Gary H. Marchant, from Shoshani & Marchant (2001)].

i-b4edf18885403d6a312c62c2277a4f55-Elephas-throat-musculature-Shoshani.jpg

Admittedly, the idea that any animal might be able to carry around an internal 'flask' of stored water, and rely on it in times of stress, is extremely bizarre. But similar abilities have been suggested for some other mammals, and the idea was first proposed back when camels were also thought to carry water in their stomachs. Given that there are published observations and data that do seem to indicate the presence of the pouch in elephants, however, this isn't the end of the story...

Next: baleen whales, reindeer, takins and quolls.

For the previous article on pouches, pockets and sacs in mammal heads, necks and chests, see...

If you're interested in tracheae and their role in respiration, vocalising and such, or on any of the associated structures in the neck, check out...

And for more on proboscideans modern and extinct, see...

Refs - -

Garstang, M. (2004). Long-distance, low-frequency elephant communication Journal of Comparative Physiology A, 190 (10), 791-805 DOI: 10.1007/s00359-004-0553-0

Shoshani, J. 1998. Understanding proboscidean evolution: a formidable task. Trends in Ecology & Evolution 13, 480-487.

- ., Agnew, D., Watson, G., Marchant, G. H. & Matsac, E. C. 1997. The pharyngeal pouch: a unique receptacle in the throat of an elephant. Proceedings of the National Conference of the American Association of Zoo Keepers 23, 74-25.

- . & Marchant, G. H. 2001. Hyoid apparatus: a little known complex of bones and its "contribution" to proboscidean evolution. In The World of Elephants - International Congress, Rome 2001, pp. 668-675.

Shoshani, J. & Tassy, P. 2005. Advances in proboscidean taxonomy and classification, anatomy and physiology, and ecology and behavior. Quaternary International 126-128, 5-20 [free pdf here].

Watson, M. 1873. Contributions to the anatomy of the Indian elephant (Elephas indicus). Part III. The head. Journal of Anatomy and Physiology 8, 85-94.

Tags

More like this

all members of Elephantida: the proboscidean clade that includes gomphotheres and elephantids, but not mastodonts or more stem-ward proboscideans

Thus saith the PhyloCode:

'Recommendation 9.3A. If a name is intended to refer to a crown clade, all of the internal specifiers used in the definition of that name should be extant.'

The only extant proboscideans are Elephas and Loxodonta, respectively; thus, defining Elephantidae is straightforward. But how, then, is Proboscidea defined? Which extinct specifier are we supposed to use?

Hmm. I haven't yet seen anyone use Proboscidea for the crown-group only, and such usage would definitely go against tradition (so... virtually all fossil elephanty-things, possibly including mammoths in some topologies, are not proboscideans under this suggestion).

Hold on - were you referring to Elephantidae? If so, did you possibly confuse Elephantida with Elephantidae?

such usage would definitely go against tradition (so... virtually all fossil elephanty-things, possibly including mammoths in some topologies, are not proboscideans under this suggestion)

Indeed; that would be an undesirable outcome, to say the least.

were you referring to Elephantidae?

Yes, the traditional 'family' which includes the extant elephants.

If so, did you possibly confuse Elephantida with Elephantidae?

I'm pretty sure that I didn't, although I must confess that I've never even heard of the taxon Elephantida until now...

Ok, thanks. If you are referring to Elephantidae, I don't think I used it to mean anything other than the crown-clade. Note that Elephantidae is - in that cladogram used above - only used for the little group at the 'tip'.

If you are referring to Elephantidae, I don't think I used it to mean anything other than the crown-clade.

No, but - just to be clear - my original inquiry was not about the phylogenetic definition of Elephantidae. What I would like to know is how Proboscidea should be defined, and which extinct specifier we should use to 'anchor' it. Should deinotheres be part of Proboscidea? What about moeritheres? Or barytheres? Or anthracobunids?

(Incidentally, if you dinosaur paleontologists think that brontodiplodocus is a hideous taxon name, please notice that there is an elephanty-thing called Stegotetrabelodon in Darren's cladogram.)

which extinct specifier we should use

Addendum: I was assuming there that the other specifier of Proboscidea must be an extinct taxon (PhyloCode Recommendation 9.3A notwithstanding) - because otherwise the taxon Proboscidea becomes de facto identical with the taxon Elephantidae.

Workers have near-universally used Proboscidea for all those taxa closer to Elephas and Loxodonta than to the desmostylians, sirenians or other tethytheres, though I can't see an explicit definition in the literature. While anthracobunids are regarded as early proboscideans by many, several studies put them as the proboscidean sister-group (e.g., Tassy 1996, Shoshani & Tassy 2005)(and, yes, I know that, historically, anthracobunids have been all over the shop).

Other than anthracobunids, all those famous archaic proboscideans of the Paleocene and Eocene are in Proboscidea. Ok, Maglio (1973) excluded moeritheres, barytheres and deinotheres from Proboscidea (thereby making Proboscidea synonymous with Elephantiformes), however, but this is very much a minority opinion. I have not seen Tassy (1988) - it might be useful.

As for Stegotetrabelodon, it's got nothing on the tortured monster that is Parapropalaeohoplophorus (named in 2007). I think this is some kind of sick joke among glyptodont workers. It's modelled on Propalaeohoplophorus (named 1887), which was itself modelled on Palaeohoplophorus (named 1883), which in turn was named as an older relative of Hoplophorus (named 1838).

Refs - -

Maglio, V. J. 1973. Origin and evolution of the Elephantidae. Transactions of the American Philosophical Society of Philadelphia, New Series 63 (3), 1â149.

Shoshani, J. & Tassy, P. 2005. Advances in proboscidean taxonomy and classification, anatomy and physiology, and ecology and behavior. Quaternary International 126â128, 5â20.

Tassy, P. 1988. The classification of Proboscidea: how many cladistic classification? Cladistics 4, 43-57.

- . 1996. Who is who among the Proboscidea? In Shoshani, J. & Tassy, P. (eds) The Proboscidea: Evolution and Palaeoecology of Elephants and Their Relatives. Oxford University Press, pp. 39-47.

If there is such a pouch, and it is used for carrying water, I suspect it would be used to transfer water from adults to offspring under circumstances where the adult could get water but the offspring could not, for example during a drought where the water table has dropped so low that only a very long adult trunk could reach it.

If so, then it might be a regional adaptation for desert elephants. In an elephant in captivity it might atrophy due to lack of use both from continuous availability of water, and also to reduced communication with other elephants.

Oh yeah (re: comment 7) - I'd not properly appreciated that you were drawing attention to Stegotetrabelodon because it looks like a hybrid of two existing generic names. There's also Stegomastodon and Stegodibelodon (Dibelodon is an old name for Rhynchotherium). Did you know that there's a sauropod called Cathartesaura?

Daedalus2u:

it might be a regional adaptation for desert elephants

But the above-cited Watson found the pouch in an Asian elephant - and they don't live in deserts.

Darren:

Did you know that there's a sauropod called Cathartesaura?

I didn't. That name is actually pretty clever.

So Elephantiformes is everything closer to living elephants than to deinotheres, barytheres, or moeritheres, while Elephantimorpha is Elephantidae+Mammutidae, making 'formes larger than 'morpha?

By Andreas Johansson (not verified) on 12 Oct 2010 #permalink

Yes, Elephantiformes Tassy, 1988 is everything closer to Elephas than Deinotherium (thereby including Palaeomastodon, Phiomia and Elephantimorpha) while Elephantimorpha Tassy & Shoshani, 1997/8 (in Shoshani et al., 1998) is everything closer to Elephas than Phiomia (thereby including Mammutida as well as Elephantida).

Maybe I'm missing something, but could it be that the elephants are simply carrying the water in their mouths? I'd guess I could hold a cup or two in mine; scaling up, a gallon doesn't seem unreasonable.

Addendum: I was assuming there that the other specifier of Proboscidea must be an extinct taxon (PhyloCode Recommendation 9.3A notwithstanding) - because otherwise the taxon Proboscidea becomes de facto identical with the taxon Elephantidae."

Only if it's given a node-based definition, and there's no a priori reason to do that. Based on Naish's comment, it sounds like a branch-based definition would be preferable, e.g., Clade(Elephas maximus â Trichechus manatus ⪠Desmostylus hesperus). Or, just make it a total clade (for which the formal name "Pan-Elephantidae" or the informal name "pan-Elephantidae" is also possible) and allow desmostylians to fall out as proboscideans (stem-elephantids) or sirenians (stem-trichechoids). (Or is Sirenia a crown clade?)

True pouch or not, how much is actually known on how elephants produce their varied vocalizations?

Mike (comment 13): if the elephants reported to be 'storing' water had been merely holding water in their mouths, I would expect them to be in close vicinity to a water source - it's not easy to imagine an animal engaging in activity for hours or even days while keeping water in its mouth. Yet elephants seen to extract water from their insides had been engaged in vigorous activity, and in cases were seen to perform this trick during times of severe drought.

Iain and Oria Douglas-Hamilton's book Among the Elephants includes four accounts of the behaviour (p. 113): in one, an adult male was waking up from sedation when it was seen to "put his trunk deep inside his throat and [suck] out water"; in another (originally told by 19th century hunter Gordon Cumming), a bull was seen to extract water after being pursued on horseback for several miles; a third case also discusses a bull chased "for some miles" by a land rover; and, in the fourth, a small calf was photographed performing the act after standing alongside its dead mother during the 1971 Tsavo drought.

Maybe the water storage is in the stomach and elephants can reach down into their stomach to extract material to feed to calves to give them the proper gut flora.

It might also be saliva. Elephants would need a lot of saliva to fully wet the dry vegetation they are eating. Being able to express saliva, extract it with their trunk, and apply it to their skin would be a good adaptation to prevent overheating. The activity after being chased for several miles suggests overheating to me.

Well, if elephant drinks a lot of water, it could vomit water and draw it using the trunk. No special anatomy needed.

One question is how regular and practically useful is that situation? I guess water quickly disappears from the stomach, so it is not avialable far from the waterhole.

Fascinating!

I seem to see a problem with the diagram ['by Gary H. Marchant and originally appeared in Shoshani (1998).'] which is that it shows the nasal passages (to trunk) closed off by the soft palate and the trachea (to lungs) closed off by the epiglottis. How then can the elephant draw stored water into its trunk?
I always have imagined that elephants drawing volumes of water into their trunk do so by 'sucking' nasally - expanding the thoracic cavity with a reduction in air pressure. To do so, the trunk - like the trachea - must be sufficiently stiff to not collapse inwards - and also, either the mouth must be sealed closed from outside air at the lips and/ or the larynx must rise and seal to the soft palate (making a continuous tube of trachea and nasal passages).
If not so, then do elephants have another way to 'suck it up'? I would be surprised if there is any ability for peristalsis in the trunk..

Maybe it's gravity not suction; maybe the trunk tip can dip down into water (wherever stored), close tightly at its end (can elephants indeed do this?), then simply lift out a volume of water?

Nit: The link to the PDF for Shoshani and Marchant (2001) is broken (needs http:// in the URL).

Iain and Oria Douglas-Hamilton's book Among the Elephants includes four accounts of the behaviour (p. 113): in one, an adult male was waking up from sedation when it was seen to "put his trunk deep inside his throat and [suck] out water"; in another (originally told by 19th century hunter Gordon Cumming), a bull was seen to extract water after being pursued on horseback for several miles; a third case also discusses a bull chased "for some miles" by a land rover; and, in the fourth, a small calf was photographed performing the act after standing alongside its dead mother during the 1971 Tsavo drought.

I'm a bit confused by what this is trying to describe -- the elephants are thirsty, and they're reaching into their mouths to suck up water to drink? Why don't they just swallow the water, then?

Or is the point that they are moistening their nasal passages and mouths with this liquid (something like a human using the tongue to moisten dry lips), and then drinking it?

The latter scenario makes sense to me, but I'd like to be sure I am understanding correctly.

By Owlmirror (not verified) on 14 Oct 2010 #permalink

By the way, an index for all of the papers presented at "The World of Elephants - International Congress, Rome 2001" (of which Shoshani and Marchant, (2001) is but one of many) is available here.

(remove "pdf/Indice_Summary.pdf" from the URL to see the web archive of the entire site)

I also have a url list that can be used to download all of the pdfs from the web archive site, if anyone is interested in that.

By Owlmirror (not verified) on 14 Oct 2010 #permalink

The trunk is prehensile, and can carry heavy loads, but does not have any bone in it, so it does have the capacity to sustain internal pressure gradients sufficient to lift those loads. The pressure needed to draw water into it is very small (~0.5 psi/ft of lift), much smaller than the pressure needed to lift an object of a few hundred pounds.

Because the trunk is prehensile, the musculature in it can be controlled to generate the pressure gradients necessary to exert force in any direction the trunk can support a load in, essentially all directions. That does indicate that the trunk can be controlled to generate a negative pressure with no access to the respiratory system.

The ability of elephants to spray water, shows that they can close off the end of the trunk and pressurize the fluid inside it.

Regarding Cathartesaura, I wonder if the similarity of Tylosaurus to the extant teleost Tylosurus (which is often misspelled...) is a coincidence.

Regarding the PhyloCode, behold Recommendation 10G:

When establishing a name for a crown clade that, under rank-based nomenclature, corresponds to a monogeneric "higher" taxon, the genus name should be converted for that clade rather than any of the suprageneric names that have been applied to it. Doing so will permit the use of the "higher" taxon names for more inclusive clades that extend beyond the crown.

Example 1. In rank-based nomenclature, the names Equisetophyta, Equisetopsida, Equisetales, Equisetaceae, and Equisetum have all been used to refer to the same crown clade, which is widely understood to include only the genus Equisetum. (Most of these names have also been used to refer to more inclusive clades that contain extinct species outside the crown.) When selecting a name to convert for the crown clade, Equisetum should be chosen. The names Equisetaceae, Equisetales, etc. are better applied to clades that are more inclusive than the crown.

This argues for using Elephantidae for the crown (as is already done) and Proboscidea for something much larger.

I think this is some kind of sick joke among glyptodont workers.

It's a sick parody of Florentino Ameghino's sick naming traditions. :-)

Maybe the water storage is in the stomach and elephants can reach down into their stomach to extract material to feed to calves to give them the proper gut flora.

You're confusing stomach and gut.

I'm a bit confused by what this is trying to describe -- the elephants are thirsty, and they're reaching into their mouths to suck up water to drink?

No. As it says right under the first picture:

Many people have reported elephants (both Asian and African) to reach into the mouth with the trunk and withdraw water. The elephants then spray this water over themselves (particularly their ears) to keep cool.

By David MarjanoviÄ (not verified) on 18 Oct 2010 #permalink

[from Darren: sorry, delayed by spam filter]

This abstract on saliva composition

http://www.ncbi.nlm.nih.gov/pubmed/3210214

Suggests to me that the liquid being observed being sprayed isn't water but is saliva. If saliva was always swallowed, there would be no need for it to be low in sodium. Calcium and phosphate are understandable for remineralization of teeth.

In sheep, it is reported that electrically stimulated salivary glands can produce as much as 30 mL saliva per gram of salivary gland per hour.

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1363181/

In the elephant, the salivary gland weighs 7.4 kg. If it can produce similar quantities of saliva, that would be 3.7 liters per minute. That is a high enough production rate that the observed flow of liquid could be saliva.

Sorry to continue this discussion so belatedly, but I had somehow managed to overlook David's posting.

This argues for using Elephantidae for the crown (as is already done) and Proboscidea for something much larger.

OK, but (as I was actually looking for a specific answer to my original inquiry about how to define clade Proboscidea) I'll have to ask again:

If Proboscidea = the least inclusive clade that contains taxon X and Elephantidae (i.e., Elephas + Loxodonta), then what is taxon X? Under the PhyloCode, is there, or is there not, a definitive answer to that question? Or is any researcher free to pick any vaguely elephant-ish extinct taxon at whim and use it to anchor Proboscidea on an ad hoc-basis?

Has nobody studied the composition of this water? I use the term water as opposed to saliva as the characteristics of an elephant's saliva are different from that of the water accessed. The water is neither acidic, nor stringy in nature. I have seen an elephant do this on frequent occasions. The elephant was of such a nature that her mouth could be fully inspected and she tended to have saliva that was thicker than average, so was easily distinguishable from the removed water.