Of Pezosiren portelli and Steller's sea cow

I don't have time for anything at the moment, it's terrible. So here's this, from one of my talks on marine mammal diversity and evolution...




For more on Pezosiren portelli, see Domning (2001); the diagram contrasting the prorastomid, protosirenid and crown-sirenian is from Domning (2000). The Steller's sea cow painting is by Maurice Wilson. Sirenians were previous covered on Tet Zoo in When manatees crossed the Atlantic.

Refs - -

Domning, D. P. 2000. The readaptation of Eocene sirenians to life in water. Historical Biology 14, 115-119.

- . 2001. The earliest known fully quadrupedal sirenian. Nature 413, 625-627.


More like this

Interesting. I wonder what sort of transitional stage sirenians went through between their terrestrial afrotherian ancestors and the highly derived crown-sirenians we all know and love. Was it some sort of semi-aquatic freshwater hippo-like herbivore, or could it have been more marine. Alternatively, it may have been semi-aquatic and freshwater dwelling, but it may not have been hippo-like.

I suppose the only way to determine for sure is to go find some more Paleocene sites in Africa.

By Anonymous (not verified) on 03 Mar 2011 #permalink

Yay for Sirenia!

Also, the body of Hydrodamalis gigas and other hydrodamalines is proportionately large relative to the head when compared to other dugongids (and trichechids), an adaptation to life in colder waters.

Anyone at all interested in fossil sirenians should take time to look at Jorge's blog - it's one of my favourites, I congratulate you :) One quick question you might like to answer, Jorge: are sirenians particularly abundant and/or diverse in the Caribbean region compared to elsewhere (they certainly seem to be)? If so, any idea why? Centre of endemism? I suspect this has already been answered in the literature but, as we all know, I'm lazy.

It's such a tragedy that Steller's sea cow was hunted down so quickly after its discovery. It would have been interesting to see how it fit into its ecosystem. I'm particularly curious about finding out how it interacted with the marine predators of the region such as killer whales or sharks. I find it hard to believe a slow and fairly harmless sirenian could properly fight off or outrun these predators to protect itself or its offsprings...

How many fossils are still around? If it was hunted by whalers, maybe they dragged the carcasses up on land to butcher them, leaving the bones on terra firma where parts of the bones may remain today. I ask, because of the success of Svante Pääbo and his team at Max Planck Institute to sequence the genomes of extinct species from old bones.
Of course, to get enough diversity for a *living* species you need at least 300 individuals (more than the number of individuals likely to be preserved as specimen in museums), which is why I ask about the numbers...

By Birger Johansson (not verified) on 03 Mar 2011 #permalink

Hi Darren,

Wow, thanks!
Yes, indeed, sirenians, particularly dugongids, seem to be more diverse in the Western Atlantic-Caribbean (WAC) region specially from the Oligocene through the Pliocene, Domning (2001) elaborates on this. So far there are about 14 species described from that region during that interval of time (plus at least 4 more that will be described in the next two years or so). During that same time interval, in the Tethys/Paratethys region, there are about 10 species, and from the Indian region 4 are known.
The seagrasses seem to have been more diverse in the WAC region, so there were enough resources to sustain three (maybe even four) sympatric species and drive speciation. There's lots of endemism, specially when talking about dugongines (which seem to have originated there). It is very likely that still more species will be discovered in the future, specially in the Greater Antilles. Even taxa that had been thought (for many years) to have originated in the Tethys region are showing up in earlier deposits in the Caribbean!

Domning, D. P. 2001. Sirenians, seagrasses, and Cenozoic ecological change in the Caribbean, Palaeogeography, Palaeoclimatology, Palaeoecology 166:27-50

How many fossils are still around?

According to Mattioli & Domning (2006) about: "27 squeletons, 62 additional skulls, and more than 550 bones"

Mattioli, S. & D. P. Domning. 2006. An annotated list of extant skeletal material of Steller's sea cow (Hydrodamalis gigas (Sirenia: Dugongidae) from the Commander Islands. Aquatic Mammals 32(3):273-288.

We just made a cast copy of a Steller's sea cow for a museum in Scotland. Such a strange animal. It's a shame the NHM in Vienna has their skeleton jammed way back in a corner.

I've always wondered about the natural history of the Steller's Sea Cow and happened to come across a nice overview of what is currently seen by some to be the state of these SeaCows back in those days, in a 2007 book "Steller's Island" by Dean Littlepage, in which the author suggests that these creatures inhabited only a couple of suitable off-shore environments on a handfull of islands which presented relatively warm water refuges following the submergence of Beringia as the current inter-glacial proceded. Previously the beringian land-mass prevented warm Pacific waters from entering the Arctic Ocean and caused those warmer Pacific waters to pool along the shores of the North Pacific and it was then that their range extended from the Shahkalin Islands to Northern California. The author suggested that at the time that the first europeans began to exploit the species they were already in reduced numbers and confined to a very limited range.
Here's a link to the amazon page.

I'm reading a novel, and I believe that the sirenians origins can be traced to the Demostylians. But, when did Steller's Sea Cow appear in history? Does anybody know? I think it's a shame that it went extinct because, like the Dodo, it is an odd animal and it's ecological niche is somewhat of a mystery.
I recommend "Neptune's Ark" by David Rains Wallace for paleo-marine biology, it's very informative.


By Harrison Ainsworth (not verified) on 03 Mar 2011 #permalink

Birger Johansson: There are still tons of sea cow bones on Commander Islands beaches.
valagos: It apparently avoided predation by staying in shallow bays protected by rocky reefs. It is unclear how it managed to colonize the islands.
doug: at the time of European discovery the Commander Islands were the only island group in the North Pacific without present or past human inhabitants. I think it's a much better explanation of the fact that the sea cow survived only there.

Haven't posted in a while, but seeing as this is a marine mammal post...

1) As far as recent bones of Hydrodamalis go - I read somewhere that most of them sitting around on Bering island had already all been picked up. Sorry, I can't provide a reference right now, but that's the situation as far as I'm aware.

2) Sirenians did not evolve from desmostylians, and the inverse is also not the case. Sirenians can be traced back into the Eocene, while Desmostylians have a fossil record only back to the Oligocene, and presumably evolved in southern or east asia, and (last I remember) potentially share some relationship with anthracobunids. Sirenians and desmostylians do share common ancestry, however, and are both marine clades of the Tethytheria.

3) Damn, I wasn't aware that there were 27 skeletons; that's quite a bit more than I remember. 62 skulls, damn.


I'm reading a novel, and I believe that the sirenians origins can be traced to the Demostylians.

You can believe that, but that's not what the evidence says. Desmostylia appears in the fossil record at least about 15 million years later than Sirenia. The divergence between the common ancestor of Sirenia-Proboscidea likely happened sometime in the Late Paleocene or earliest Eocene, whereas that of Desmostylia-Proboscidea (or Anthracobunidae) by the Late Eocene or Early Oligocene. So, like Bobby says, one cannot be the direct ancestor of the other.

Speaking of desmostylids...what was their M.O.? I have a paper in my archives that suggests they were doggie-paddlers in the way of modern bears, and that they grubbed up marine flora with their tusks. But the paper is quite old. Any news on these strangest of potentially semi-aquatic herbivores?

Boesse: in the late 1980-s finding a bone took about half an hour of walking on the beach. I would be surprised if the situation has changed that much. If you are interested, I can ask a friend of mine who works there every summer.

Re subfossil bones on Bering Island: there may be a lot more than the ones you can easily pick up. Nordenskiöld who was there in 1879 and collected quite a lot of bones noted that the best way to find them was to probe 1-2 feet deep in the beach ridges. However he noted that ribs were scarcer than other bones because the locals used them as runners for sledges!

By Tommy Tyrberg (not verified) on 03 Mar 2011 #permalink


I ask, because of the success of Svante Pääbo and his team at Max Planck Institute to sequence the genomes of extinct species from old bones.

Steller's sea cow mitochondrial DNA was extracted already in the 1990ies (Ozawa et al., 1997) from bones collected on Bering Island. (I am not aware of any more recent molecular studies of this species, however.)

Ozawa, T., Hayashi, S. & Mikhelson, V.M. 1997. Phylogenetic position of mammoth and Stellerâs sea cow within Tethytheria demonstrated by mitochondrial DNA sequences. Journal of Molecular Evolution 44, 406-413.

@5 Birger: you may be able to buy the bones from Steller Sea Cows. A company called Dancing Man Knives in Alaska is selling knives with handles reportedly made of that bone, and they have been for years (http://www.local-showcase.com/Dancing_Man.html). I don't know where they're getting it or if its real, but it can't hurt to talk to them. Their ulus are good tools, too.

As for sirenians and desmostylians, my take was that sirenians are to seals as desmostylians were to sea-lions. That might be overly simplistic, but my uneducated thought was that desmostylians seemed to be more agile, and perhaps adapted to high energy environments, such as the kelp beds of the Pacific, estuaries, and similar environments that demand a modicum of athleticism to keep from being smashed by the waves. Sirenians seemed adapted more for efficient cruising than at agility.

By heteromeles (not verified) on 04 Mar 2011 #permalink

I like that last comparison! It also seems reasonable to me. A few questions regarding desmostylians:
- Were desmostylians truly marine or were they more semi-aquatic and coasthugging creatures? Their absence beyond the northern Pacific might be an indication there.
-On the other hand, is it likely that they also occurred in the southern Pacific or even the Indian Ocean?
-Also, do we have any idea why demostylians went extinct?
-I recall demostylians being depicted with very short trunks (or at least elongated noses). Is this realistic?
- Do we have any clues about the natural history of these creatures, beyond their habitat and diet?

It appears that I was wrong then about the abundance of bones - good! I'm glad I am, in this case. I wonder where I had heard that...

RE: Desmostylians -
Oxygen/carbon/strontium isotope data places desmostylians as feeding within somewhat fresh water (or less saline water anyway). The Delta O18 is marine, while the Sr isotope signal appears terrestrial. So, whatever they were doing, they were doing it close to the shoreline. That being said, desmostylian fossils have only been found in marine deposits, and never in terrestrial deposits (although it is important to recognize that many of the terrestrial deposits in California are pretty far inland - then again, so are some marine desmostylian-bearing deposits).

Last I recall, desmostylians are thought to have fed upon algae or kelp; otherwise, their posture during terrestrial locomotion is thought to have been pretty bizarre (Domning 2002), with a digitigrade, chalicothere-like stance for its feet, a near vertially oriented pelvis and habitually dorsiflexed vertebral column. Additionally, a paper by Phil Gingerich (2005) indicates that in terms of swimming style and skeletal proportions, Desmostylus was most similar to the polar bear (which is not surprising, looking at the skeleton of desmostylians). I don't think there's much evidence or need to invoke a trunk for desmostylians.

Otherwise - they disappear sometime during the late Miocene, and some of their youngest fossils (in California, anyway) are ~10-12 Ma. This is roughly the same time that hydrodamaline sea cows start getting gigantic, and there is quite a bit of turnover in the NE Pacific marine mammal fauna from the Tortonian into the Messinian. I can't recall of the top of my head, but there have definitely been some hypotheses proposed - perhaps Jorge can help me out here, or maybe Brian Beatty will see this.

I'm intrigued by the diagram next to the Pezosiren dot-point, suggesting that the relative positions of the Center of mass and Center of bouyancy have changed over the course of Sirenian evolution. This is plausible enough, but are the remains of Pezosiren and other Eocene Sirenians complete enough to allow calculation of their Centers?
I've also had a niggle for years about Moeritherium. The most recent cladograms I'be seen put it closer to extant Proboscidea than to Sirenia or Desmostylia, but its apparently aquatic life-style makes it seem like a good analogue, at least, for the ancestors of the Sirenia. (And the description of the likely life posture of Desmostylia in post #20 sounds like what Embrithopoda might have evolved toward if they'd had a chance!) Prediction: new discoveries will lead to major revision in our picture of the early evolution of the Thethytheria!

By Allen Hazen (not verified) on 04 Mar 2011 #permalink

I thought Brian Beatty had proposed that their extinction might relate to marine productivity moving farther offshore beyond the foraging range of Desmostylians. Alternatively, I am just imagining that.

I think it's looking increasingly likely that the common ancestor for elephants and sirenians was aquatic. However, IIRC, stable isotope analysis actually supports embrithopods being terrestrial (or at least as terrestrial as modern elephants and such).

By Morgan Churchill (not verified) on 05 Mar 2011 #permalink

Morgan Churchill (#22)-- Thanks, didn't know about the stable isotope stuff. I was going on the tradition of interpreting various skeletal features of Arsinoitherium as suggesting aquatic tendencies.

By Allen Hazen (not verified) on 05 Mar 2011 #permalink

I'm not sure if its entirely clear why the desmostylians went extinct. It does however correlate with the proliferation of hydrodamalines. Were desmostylians outcompeted by hydrodamalines? Or did they went extinct because of increase in cold climate drove the seagrasses away from their range? I really don't know.

Were the desmostylians eating sea-grasses? Sorry for the question, but I didn't realize that one could tell the isotopic difference between sea grasses (angiosperms) green algae, and kelps.

I guess I was thinking about desmostylians eating kelp. Being on the west coast of the US, one thing I do know is that kelps fluctuate enormously both within the year and between years. They're fast growing, but they also get shredded by storms, as does sea grass.

In such a seasonal environment, I'd expect any grazer to have to both endure strong seasonality (summer feast, winter famine) and have to move as their food plants changed. Manatees in the Amazon have to deal with this type of environment now.

Since desmostylians were big, I'd guess that simple environmental stochasticity drove them extinct. They may have simply gotten caught in a situation where they couldn't find big enough patches of food plants to allow them to reproduce, or they may not have been as good at surviving bad seasons as the hydrodamalines. Obviously I don't know either, but I do know that the north Pacific definitely does not live up to its name. It's a very unstable place.

By heteromeles (not verified) on 06 Mar 2011 #permalink


Yes, desmostylians were eating seagrasses and yes, you can tell the difference between seagrasses and kelp. It is even possible to tell, in the case of sirenians, if they preferred large or small seagrasses, which I think its really cool!

Clementz, M. T., K. A. Hoppe & P. L. Koch. 2003. A paleoecological paradox: the habitat and dietary preferences of the extinct tethythere Desmostylus, inferred from stable isotope analysis. Paleobiology 29:506-519.

Clementz, M. T., S. Sorbi & D. P. Domning. 2009. Evidence of Cenozoic environmental and ecological change from stable isotope analysis of sirenian remains from the Tethys-Mediterranean region. Geology 37:307-301.

Thanks Jorge,

Someone called seagrasses "botanical whales," so there's some interesting parallelism here.

Right now, seagrasses are rare along the California coast, and their biomass appears to be strongly seasonal (based on my frequent walks near a seagrass bed, and the amount that washes up over time. I'd still hypothesize that contraction of seagrass range may have doomed the desmostylians.

I guess the question is, what were the Miocene hydrodamalines eating? I know that manatees eat seagrass, but steller's sea cow reportedly ate kelp. Were the hydrodamalines simply more catholic in their dietary preferences?

By heteromeles (not verified) on 07 Mar 2011 #permalink

Indeed, hydrodamalines seem to have been eating mainly kelp, hence the trend towards reducing the dentition (to none in Hydrodamalis spp.) and rostral deflection.

So...perhaps dietary differentiation is a viable hypothesis for the turnover between Desmostylians and Hydrodamalians?

By heteromeles (not verified) on 09 Mar 2011 #permalink

Previously the beringian land-mass prevented warm Pacific waters from entering the Arctic Ocean and caused those warmer Pacific waters to pool along the shores of the North Pacific and it was then that their range extended from the Shahkalin Islands to Northern California.

Hard to believe that only the Commander Islands maintained warm enough water for this species to survive--warmer than northern California?--especially when given the proven ease with which the sea cows were hunted to extinction. I can totally see the Commanders as a last refuge for their population--from people, all too temporarily.

Of course, to get enough diversity for a *living* species you need at least 300 individuals (more than the number of individuals likely to be preserved as specimen in museums), which is why I ask about the numbers...

That is not a hard-and-fast rule, there are many examples of viable populations being established from fewer individuals.
If you're speculating about recreating a species from DNA, trying to clone hundreds of separate individuals would probably not be the way to go. Better to assemble the best possible consensus genome and fiddle with MHC genes and whatever is likely to affect viability (but that assumes some greater hurdles can be cleared in this SF scenario, such as where to find a womb to incubate these things!)

Re: Windy (#31), "such as where to find a womb to incubate these things!"

I know Steller's Sea Cow is bigger than extant Dugongs, but (assuming nothing else goes wrong) could a Dugong carry a Steller's fetus, not to term, but at least to a viable stage of development (caesarian delivery possible if a premature but viable Steller's infant is a bit larger than a typical full-term Dugong)?

By Allen Hazen (not verified) on 13 Mar 2011 #permalink


could a Dugong carry a Steller's fetus, not to term, but at least to a viable stage of development

Before we could attempt to use a dugong as a surrogate mother, we would first need to know how to successfully breed dugongs under captive conditions; AFAIK, dugongs have never reproduced in captivity (and only a handful of individuals have ever even been kept in zoos and aquaria).

Manatees, on the other hand, have been bred in captivity - but they are of course far more distantly related to Hydrodamalis than the dugong is.

@Vladimir Dinets #11: Interesting! If they were confined to the Commander Islands because they had been hunted out everywhere else, then the present potential habitat of the sea cow might extend all the way to California. Presumably it would be somewhat patchy. Perhaps the sea cows would migrate seasonally between sheltered bays where they could escape the winter storm season and the more exposed coasts where the kelp forests grow. I wonder whether they could coast along on their fat reserves for a long period like migratory whales.

Fjord country might be ideal for such a creature. Imagine winter herds of sea cows like so many knobbly half-sunk logs in the quiet waters of the inner parts of Puget Sound.

I wonder how they avoided orcas off the Commanders. Or anywhere else for that matter.

By Jenny Islander (not verified) on 15 Mar 2011 #permalink

My follow-up post went poof TWICE. ARGH.

In summary: Here is an English translation of Steller's work on the sea cow. http://digitalcommons.unl.edu/libraryscience/17/ He based it on almost a year of continual observation of the animals in the wild.

I infer from his account that the population in the Commander Islands was in marginal habitat. The animals did most of their eating in North Pacific kelp forests, but hung out mainly at river mouths. They were killed regularly by sea ice and by storms coming in off the Bering and they were visibly emaciated at the end of the winter. IMO, this implies that they would have done better (bred more successfully, grown bigger, grown more quickly, lived longer?) someplace with estuaries and kelp that was more sheltered and also warmer, possibly with more plant species. This would put their largest areas of present potential North American habitat in a range from the Inside Passage to Puget Sound and south to the Golden Gate. Their refugium during glacial intervals might be off Baja California, or for all I know the coastline that is now underwater provided more habitat. Distribution on the opposite coast of the Pacific would be similar, with some island herds in marginal habitat between the two main ranges.

By Jenny Islander (not verified) on 16 Mar 2011 #permalink

I'm sorry I missed all this discussion, especially the part about Desmostylia. I think all the salient points were made, though I would urge that a distinction between algae eating and seagrass eating is important to make. As is the difference between what isotopes and other methods can discern about diets. Animal diets are unfortunately very plastic and complex, as is the nature of the data we use to study it (isotopes, microwear, morphology) - we all need to be cautious about understanding the nature of the data we are using. That anyone can be certain about hypotheses about competition between hydrodamalines and desmostylians without really statistically knowing what they ate or how their temporal and spatial distributions overlapped should make us all uncomfortable. Moreover, modern competition studies start with systems like that and then study how resource limitation, if it exists, may play a part. We haven't established any notion about how algae or seagrasses were limited over the period of time of hydrodamaline/desmostylian spatial and temporal ranges, so how can we know it was competition?

Sorry for the rant... now you see why I get things done at a snail's pace. There are just too many variables to control for, leaving us just to wave our arms and hope people are convinced, even if we aren't.

Going back to the post's title - I was hoping to see more discussion about Pezosiren...
I'm working with Daryl Domning right now on that critter's enamel microstructure, and noticed that virtually nowhere on Scienceblogs is there any mention of enamel microstructure (aside from one post by Afarensis on humans, with no images of it there).
I wonder, why not? There is some REALLY cool work on what amounts to the evolution of histology, and I bet lots of people would really get into it.

Thanks for comments, Brian. I would be happy to discuss Pezosiren at length (and desmostylians), given appropriate time and perhaps access to new images and data :) As for enamel microstructure... histology of any kind doesn't exactly push my buttons, but I know how important it can be as goes biology and evolution.

Re #37 and #38--
Enamel microstructure is one of the key pieces of evidence about Mesozoic mammals (hey, if all you've got is "the tooth, the sometimes broken and incomplete tooth, and nothing but the tooth" you look at it as closely as you can!). The down side (my impression from reading) being that there are a limited number of possible microstructures, with the derived ones having evolved repeatedly all over the family tree!
But I'll be interested to see what it tells us about Pezosiren.

By Allen Hazen (not verified) on 24 Mar 2011 #permalink