Laelaps is the blog of 
Last week I wrote about the shuffling and reshuffling of relationships between whales, hippos, pigs, and an extinct group of mammals called raoellids. One aspect of the paper I did not comment on, however, was the problematic placement of the enormous predator Andrewsarchus.
In November of 1924 Henry Fairfield Osborn described the huge skull of a predatory mammal discovered during a American Museum of Natural History expedition to Mongolia. Unfortunately there was little other than the skull left of the great beast, but Osborn thought it belonged to a gigantic omnivore. Osborn named it Andrewsarchus and stated that it was a kind of extinct hoofed predator called a mesonychid. It was this group of mammals that, until very recently, were considered to be the ancestors of whales.
More than 80 years later sparingly little is known of Andrewsarchus. Despite an appearance in the BBC's Walking With Prehistoric Beasts and a small amount of popularity the skull at the AMNH* still represents almost everything we know about this animal. Some things have changed, though. It is possible that Andrewsarchus is not a mesonychid but belong to a related group, the triisodontines. (Alternatively the triisodontines could fall within mesonychia.) It once seemed fairly "secure" within the mesonychia but this is not so anymore.
*[I posted a photo I had taken of that skull on my old blog in 2007, and I was surprised to find it attributed to someone else on Wikipedia. I have left the image up but changed the attribution. I am glad people have found my photos to be useful but please, if you're going to use them, attribute them correctly!]
Indeed, Andrewsarchus also poses a problem to systematic analysis of the relationship between whales, hippos, and other hoofed mammals. So little is known of it that it truly has the potential to throw a monkey wrench in attempts to construct phylogenetic trees, as was the case in the paper by Geisler and Theodor. The appendix to their recent communication shows that when Andrewsarchus was included the result looked like a short lawn of taxa with little resolution. When it was removed the results were much improved.
It seems that Andrewsarchus occupies an unstable position. Is it a mesonychid, or is it some kind of artiodactyl more closely related to hippos and/or whales? No one can say for sure based upon the present evidence. Even though the skull of Andrewsarchus is informative we truly need to find a complete, or nearly complete, skeleton to better understand just what sort of creature it was. Until that time it will remain something of an enigma.
Life Science
Comments
It would be nice if we had more remains of this creature besides it's enormous skull, so we could know what the rest of the body looked like, and where the heck it's affinities lie. "Trisodontines"? That's a new one on me, I'll have to look it up!
Posted by: Raymond Minton | March 24, 2009 1:51 PM
Thanks, Raymond. Admittedly I do not know much about the triisodontines at present, but when I have a better handle on them I will write more. If you can find it the best resource to look at might be Ken Rose's The Beginning of the Age of Mammals.
Posted by: Laelaps | March 24, 2009 2:15 PM
Your image has been nominated for deletion after this post, some guy had uploaded it with false copyright information. That happens once in a while. This image will be used instead: http://commons.wikimedia.org/wiki/File:Andrewsarchus_mongoliensis.jpg
Posted by: Michael | March 24, 2009 6:24 PM
Michael, please check with the AMNH. I believe it has a standing policy: no photographs of Museum exhibits may be published anywhere without the Museum's permission. At least, that's what I was told several years ago when I wanted to use some pictures from the Museum on a webpage.
Brian, a nice post. The Walking with Beasts animation isn't very convincing, but then as I recall, not much about that show was. I didn't know that Andrewsarchus was still just the skull and little else. Rather annoying. I wonder where it does fit in the mammal family tree?
Posted by: wolfwalker | March 24, 2009 8:17 PM
"Wherefore" = "why".
Posted by: Blake Stacey | March 24, 2009 10:23 PM
Blake; Thanks. I actually knew that and was feeling a little meta about Andrewsarchus. (akin to "Why is the platypus?") If we knew where to find some more we might be able to get at that question. :)
Posted by: Laelaps | March 24, 2009 10:38 PM
Ah. That makes more sense.
Posted by: Blake Stacey | March 25, 2009 12:40 AM
One thing that bothers me about the Walking With Beasts reconstruction of Andrewsarchus (and most other reconstructions I see) is that the eyes are placed pretty close to the top of the skull, like a dog's. But if I'm interpreting the skull correctly (and remembering something I think I read), the eyes of Andrewsarchus are placed rather freakishly low on the skull -- closer to the jawline than the peak of the sagittal crest, it looks like.
See, for example, the reconstruction of Andrewsarchus in the Wikipedia article about same -- I think its eyes are placed correctly.
Am I right? Or am I, as an amateur, completely misinterpreting the skull?
Also, I recall once reading on the Dinosaur Mailing List that fish-hunting waterbirds like herons tend to have their eyes set lower in their heads -- this gives them a better view of fish caught in their long bills so they can better secure their hold or reposition the fish for swallowing. If this is true -- and frankly, after looking at some photos of birds' heads I can't really tell if it consistently is -- I wonder if the low placement of the eyes in the skull of long-jawed Andrewsarchus migh reflect a fishing lifestyle. Andrewsarchus might have lived near bodies of water and behaved something like a cross between a heron and a salmon-hunting bear.
Posted by: Stevo Darkly | March 25, 2009 2:57 AM
Wolfwalker, as far as I know about American copyright laws, that would only apply to art, such as a model of the animal or similar, for actual specimens this does not apply, as they are not man-made art, obviously. Similarly, a Zoo would not be legally able to prevent you from publishing pictures of their animals.
Posted by: Michael | March 25, 2009 8:07 AM
Stevo is right--the eye would sit ridiculously low on the skull.
Posted by: Zach Miller | March 25, 2009 4:56 PM
The name of this species indicates that the expedition was the one lead by Roy Chapman Andrews. It was funded by the AMNH. A long time ago, circa 1950, I read Andrews account of this expedition, IIRC: "On the Trail of Ancient Man." Well, they were looking in the wrong place for "ancient man," but it was an exciting book, and influenced me in my decision to make science my profession.
Bob
Posted by: Bob Carroll | March 26, 2009 6:49 PM
Andrewsarchus mongoliensis is a strange beast, not only the long rostrum, but the supraoccipital and occipital condyles seem so proportionately small when compared to the rest of the skull that it makes me wonder how it could keep its head up.
Posted by: Jorge Velez-Juarbe | March 29, 2009 10:10 PM
I once read in one book that someone actually considered Andrewsarchus to not be an actual taxa at all, but a deformed skull of "Baluchitherium" (now Indricotherium or Paraceratherium, depending on who you talk to) or vice versa. While I realize that this is foolish and unfounded speculation, and heavily suggests that whoever wrote that book had no knowlege of Andrewsarchus' anatomy, it does drive home exactly how enigmatic the classification of Andrewsarchus is.
Posted by: Metalraptor | March 30, 2009 10:00 PM
> Andrewsarchus might have lived near bodies of water and
> behaved something like a cross between a heron and a
> salmon-hunting bear.
A gigantic, long-snouted piscivore? Sounds like the mammalian equivalent of a spinosaur ;)
Posted by: johannes | April 3, 2009 3:38 AM
To me, that skull looks like some kind of crocodile type animal skull. Why do they think that it is mammal? But anyway, I agree that we need more evidence about this animal, I m more than ready to leave to mongolia to look more fossiles about this...
Posted by: dorkalias | April 20, 2009 2:07 AM
dorkalias: they "think" it's a mammal because on a basic level (ie. looking at the shape of the bones and not just the proportions), it's extremely similar to known artiodactyls like Pakicetus, Kutchicetus, Ambulocetus, Rodhocetus, etc... say, they did say it was found in a shoreline environment, right? Maybe they should have named it "Andrewcetus"...
Compare it to, say, Baurusuchus (a land-dwelling crocodile) and you'll see they really don't look much like each other at all except for the long snout.
Posted by: Gwen | May 21, 2009 12:53 AM