Neckless Wonders: Feeding Techniques of Dimetrodon and other Early Terrestrial Tetrapods

There’s been a lot of talk recently on David Peters’ blog “The Pterosaur Heresies” ( on early synapsids – vertebrate tetrapods with one hole in the skull behind the eye. I find this interesting since, as you can tell from the website’s title, Mr. Peters has a particular interest in pterosaurs – those flying Mesozoic reptiles that are often mistaken for dinosaurs.

Another creature that’s often mistaken for a dinosaur is Dimetrodon. If you’re an adult reading this, you may not recognize the name, but you’ll probably know it when you see it. Ask any 6 year old child, and he or she will immediately tell you what one looks like. It measured ten feet long, walked on four legs, had a distinctively-shaped head with almost S-shaped jaws filled with an impressive array of teeth, and its most obvious feature was that it had a huge sail on its back, probably for regulating temperature.

Dimetrodon grandis, one of several Dimetrodon species. © Jason R. Abdale. March 28, 2016.

If you want some more detailed info, Dimetrodon, meaning “two long teeth”, was a multi-speciate genus of carnivorous terrestrial synapsid amniote tetrapod, specifically a member of Spenacodontidae, which lived during the early Permian Period.

I can already tell some of you are going “Huh???” Let me see if I can explain this in regular English.

The name Dimetrodon means “two long teeth” in ancient Greek; it does NOT mean “two kinds of teeth” as you will sometimes see in books (then the name would be Dimorphodon, a name which is already used for a pterosaur – a nice David Peters segway there). The name comes from the two impressively long canines that it had on each side of its upper jaw.

Dimetrodon is the genus name; a genus is a group of related species. To be “multi-speciate” means that this particular genus contains many species. Examples of Dimetrodon species include D. grandis, D. limbatus (probably the most common species), and D. milleri. The exact number of Dimetrodon species varies, depending upon who you ask, because some people claim that certain names are invalid. The last time I checked, there were about fifteen or so different species spread out across the Northern Hemisphere from Texas, USA to Germany.

“Carnivorous” means “meat-eating”, and “terrestrial” means “lives on the land”, but I think nearly everybody knew that already.

As stated earlier, a “synapsid” is a group within a larger group of animals called terapods. The word “tetrapod” is Greek for “four feet” – tetrapods are animals which have or at one time had four limbs. These include amphibians, reptiles, mammals, and birds – wings count as limbs. Even whales and snakes are considered tetrapods because they are descended from animals which DID have four legs. Tetrapods are further sub-divided into amphibians (frogs, toads, newts, slamanders, etc.) and amniotes. An “amniote” is an animal which reproduces by laying eggs with hard shells which retain moisture. Inside the egg is a fluid-filled membrane called the “amnion” or “amnios”, which helps keep the developing embryo inside hydrated. While amphibians must lay their eggs in water to keep them moist, amniotes have the freedom of laying their eggs on dry land.

Amniotes are broadly catagorized into four groups depending upon how many holes they have in their skulls behind the eye socket and where those holes are located.

1) Anapsid – Greek for “no opening”; aside from the nostril and the eye socket, there are no other holes in the skull.

2) Synapsid – “fused opening”; one hole behind each eye, positioned low. All mammals, including humans, are synapsids.

3) Euryapsid – “wide opening”, one hole behind each eye, positioned high. Includes marine reptiles like nothosaurs, plesiosaurs, and ichthyosaurs. All euryapsids became extinct at the end of the Mesozoic Era along with the dinosaurs. NOTE: I’ve heard that the term “euryapsid” is paraphyletic, meaning that it’s an artificial group composed of many different kinds of animals that aren’t actually related to each other. But determining whether or not a group is paraphyletic is a subject for another day, and I won’t get into it here.

4) Diapsid – “two openings”; there are two holes in the skull behind each eye socket. Dinosaurs were diapsids.

Dimetrodon belongs to the synapsid group of amniotes, the one which includes mammals and the ancestors of mammals. So, in an extreme way, despite its lizard-like appearance, Dimetrodon is our great-great-great-great-great-great-great (and so on) ancestor!

In terms of how many groups of synapsids there were/are, there were/are many. One of them was a group called the “pelycosaurs” – you will often see Dimetrodon being referred to as a member of this group. The pelycosaurs first appeared at the end of the Carboniferous Period about 300 million years ago, but they really became dominant during the early part of the Permian Period, about 280 million years ago. There were many different species, but the ones which grab everyone’s attention are the sail-backed ones like Dimetrodon and Edaphosaurus.

The pelycosaurs were divided into four families, and one of them was called Sphenacodontidae, defined as a family composed of the genus Sphenacodon and all other animals that are more closely related to Sphenacodon than to other animals. The various genera that compose this group are generically referred to as “sphenacodonts”. Dimetrodon was a sphenacodont.

There, understand now?

Dimetrodon is often mistaken for a dinosaur merely because it’s prehistoric and looks dinosaur-ish. I want to show you a picture to see what I mean – it’s a pencil drawing by Vladimir Nikolov made in 2010. I’m not going to put the picture here because I don’t have permission from the original artist to publish it here. The picture is found on DeviantArt, but a word of caution – DeviantArt is known to have viruses, so I would highly suggest that instead of clicking on the link, you would instead google the image “Dimetrodon grandis Vladimir Nikolov”, and see what picture pops up in your search results. It’s a very good picture, nicely done from an artistic standpoint. I like the fact that, unlike the vast majority of paleo-artists out there, this person chose to show Dimetrodon in an active Komodo Dragon style way, rather than dragging its belly using four very scrawney limp legs. I also like the color scheme.

But there are three problems with this picture. First, I don’t think that Dimetrodon had scales of any sort. I read somewhere that preserved skin specimens actually show that it had tough leathery hide. But then again, if Dimetrodon really was a very ancient proto-mammal, you wouldn’t expect it to have scales, right? Second, he says it’s supposed to be Dimetrodon grandis, but based upon the shape of its skull, it looks more like a Dimetrodon limbatus to me – that’s the species that you’re going to see on display in most museums. Third, which is I think the most serious, Dimetrodon didn’t actually have a neck! If you were to look at any of the hundreds of Dimetrodon skeletons which are on display all over the world, you will very quickly notice that their skulls seem to be bolted directly onto their shoulders with no neck in between.

Now that I think about it, there are many examples of land-dwelling tetrapods  from the Carboniferous and Permian Periods which do not have necks, especially the pelycosaurs. I’ve seen skeletons of many pelycosaurs, either in real life or images in books and on websites, and I can safely say that NONE OF THEM HAD NECKS!!!

The absence of a neck must have had some very peculiar effects on how these animals moved and especially how they ate. A neck is a wonderful thing. A neck is a flexible apparatus which enables the creature that has it to do several things:

1) The neck contains the vocal cords. The vocal cords attached to the larynx of a longer neck will likely produce different sounds than a shorter neck.

2) The neck enables the head to move without moving the body. A head can twist and turn to see and to find food while the body can remain motionless.

3) Connected to #2, a neck enables the head to have a wide range of movement during the actual process of feeding. A carnivorous animal can chomp down on a carcass, and then twist and turn its neck so that its head can wrench off a large chunk of meat.

There’s no denying that Dimetrodon had some very impressive choppers – after all, it’s named after its teeth. It also had a very large solidly-built head in proportion to its body size, which must have given it a very hefty bite. However, the absence of a neck must have meant that it fed in a certain way. It could not attack the way that meat-eating theropod dinosaurs could with their elegant S-shaped necks, nor could they even attack the way that lizards or crocodilians (who also have short necks) could. Even crocodilians have longer necks than Dimetrodon and its Paleozoic contemporaries. Due to its large sail, a Dimetrodon certainly couldn’t “death roll” the way that a crocodile can (watch any episode of The Crocodile Hunter to find out what a “death roll” is).

Dimetrodon could probaby swing its head from side-to-side reasonably well to a certain degree, but it would have had limited up-down flexibility, maybe just enough to flick its head back a bit and gulp a piece of meat down. If Dimetrodon had a free fully-mobile tongue the way that some reptiles and mammals do like us, then swallowing food would be easier. However, if it had a fixed tongue like fish, amphibians, and crocodilians, then it would have to rely on flinging the food into the back of the mouth in order to swallow it.

Here’s how I hypothesize Dimetrodon fed. I think that once it got its jaws firmly into something, it either shook its head from side-to-side like a shark, or it could have braced the meat with its front paws and then pulled its entire body backwards, thus pulling a piece of flesh off of its prey. This would be an example of “puncture and pull” feeding. It certainly did not have the necessary head mobility or the height to enable it to do “hatchet attack” feeding, where the jaws are stretched as wide open as they can possibly get and then the upper jaw is literally slammed downwards onto the prey like a guillotine.

I am not particularly inclined to engineering, physics, or mathematics, so somebody out there has to do a 3D computer model study on skull and neck mechanics in order to replicate how Dimetrodon and its neckless ilk might have eaten. However, based upon observations of the bones alone and looking at how modern animals feed, I’m pretty sure that this is how the neckless wonders of the late Paleozoic would have eaten.

Categories: Paleontology, Uncategorized

Tags: , , , , , , , , , ,

3 replies

  1. I have been working on the skeleton of Dimetrodon milleri in its cleaning and preservation. I also have been using detailed photographs supplied to me from the Harvard University Museum of Comparitive Zoology and the University of Texas.
    I can assure you that physically Dimetrodon has a neck represented by at least 7 cervical vertebra similar in number in many mammals. They are larger in over bulk than the pre-sacral in the vertebra column that form the sail and they have elongated processes for the cervical ribs. Furthermore the dorsal spines begin to start rather robust from the axis vertebra up to the seventh in the series. When you consider the highth of the anterior sail and the depth of the cervical ribs it appears to have a rather short neck when portrayed in flesh restorations. Regardless there is a distinct distance from the occipital process to the scapula. Its certainly not long as in camels and sauropod and other long necked species but it just does not stand out as distinctly due to the sail and it thickness in life like illustrations.
    David P Letasi

    • Greetings, and thank you for your information concerning Dimetrodon anatomy. I based the illustration which you see in this article, as well as the information in the article itself, on numerous Dimetrodon skeletons which I have seen, and I stand by my claims. As I stated in the article, I have examined Dimetrodon fossils with my own eyes, and I poured over hundred of images of Dimetrodon skeletons that are on display elsewhere throughout the world. In addition to this, I have also examined the skeletons of other Permian Period tetrapods such as Sphenadocon, Edaphosaurus, Eryops, and Seymouria, either in-person or in photographs. I was already aware that Dimetrodon and other sphenacodonts did possess cervical vertebrae, and thus it “technically” had a neck, but that wasn’t the point that I was trying to make here. Dimetrodon’s neck was so short, and possessed so little flexibility due to the presence of its large shoulders and the massive amounts of flesh surrounding the cervical vertebrae, that it might as well have not had any neck at all. If you were to see a living Dimetrodon, would you actually see a neck physically separating the head from its shoulders, or would it look as though the head was attached directly onto the shoulders with no neck in between? I’d argue for the second option. As you can see from the numerous Dimetrodon skeletons which we have, and ones which you yourself have worked on, there is its head, a very short stumpy neck, and then its shoulders. The shoulder girdle is positioned so close to the back of its lower jaw that I’m amazed that it could even open its mouth without hitting it. Thus, to the human eye, a living Dimetrodon would appear as though it didn’t even have any neck at all. Its shortened neck and the large quantity of tissue which encompassed it must have severely restricted the movement and mobility of the head, and therefore it must have had an impact on how it and other similarly-built creatures must have fed. This was the point which I was trying to make.

  2. Jason.
    Very interested comments concerning the physiology of Dimetrodon. I’ve searched for quality photos of articulated specimens in the field. Every photos so far I’ve uncovered are very disarticulate and I haven’t found one showing an actual specimen where the scapula lies in relation to the skull and containing a cervical series. Every specimen I’ve examine are from highly disarticulated individuals. Even Bakkers Wet Willy has mostly mid axial articulation and disarticulated leg elements. The one I’m working on is 60 percent complete but was recovered highly fragmented and very disarticulated. I reviewed the collections at the University of Tesas paleontology lab with Dr. Chris Sagebeil reviewing their extensive Dimetrodon material. Little articulated material is represented. I also have reviewed the University of Michigan’s extensive Dimetrodon collections collected by E.C. Case in the 1930s.
    I noticed that even the reconstruction mounts on display were composed of highly fragmentary material and I feel that many specimens are physically reconstructed with a high degree of conjecture. If you have any photographs of articulated specimens in the field that could demonstrate your view point on these animals physiology alignment between the head and the position of the front shoulder and legs alignment. That would enhance you pposition greatly. I see in numerous mounts at various museums in North America that each skeletal mount varies some what in relation to the mounting of antrrior shoulder elements related to the skull and its position in the anterior portion of the rib cage and dorsal spines. This variability make some mounts look fairly longer necked while other look some what shorter necked. Regardless the cervical vertebra are articulated with ample cartilage between centra to allow a great deal of flexibility. I can see Dimetrodon to have a laterl movement of its head to at least a nearly 160 plus degree in flexibility from side to side. In fact I’m developing a mounting stance on my specimen to have a nealy 80 degree plus lateral tilt of its head and axial skeleton.
    I hope you further research your studies as I would be very interested in what you develop as the physiology on Dimetrodon has very little research. I also have worked on the range of motion on the scapula and humerous to determine the flexibility of its agility to stand fairly high off the ground. In my experimentation of determining the range of motion on these joint prosees. I determined that the Dimetrodon humerus extended beyond 45 degrees with the scapula the joint would begin to disarticulate. After looking at Dr. Scott Hartmans illustrations of a walking Dimetrodon skeleton I noticed that his legs were over extended based on my evidence. He agreed and has considered reducing its high walking stance. Even though the stance is some what lower it could still could have raised its torso off the ground just like foot print trailways have demonstrated in the Permian fossil record.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: