The Ancestry of Mammals: A Profile of the Synapsida

Amniotes can be split into two major groups: the Sauropsida and the Synapsida. Synapsids are amniotes with only one large temporal fenestra (the lower one), as seen in the highlighted skull above (Benton, 2004). The sauropsids can have none (anapsid, e.g. turtles) or two (diapsid, e.g. lizards, snakes, dinosaurs). The fourth diagrammed skull with one small temporal fenestra is a euryapsid, and is found only in some extinct marine sauropsids.

Other characteristics of the synapsids include:

  • 1 maxillar canine (or, better said, canine-like tooth), as opposed to a homogeneous bite;
  • Maxillar and quadratojugal bones meet in the cheek;
  • Narrow vertebral arch (where the nerves pass through to the body).

Synapsids split off from the rest of the amniotes during the initial amniote radiation in the swamps of the Carboniferous, with the earliest fossils from the lowest Late Carboniferous (Pennsylvanian, 310 Ma) showing the emergence of the herbivorous Edaphosauridae, one of those famous dinosaurs with the sail on their backs. Readers with an awesome childhood might know these as “pelycosaurs“, but this is a paraphyletic grouping containing the edaphosaurids and several other early “sailed” synapsids, such as the carnivorous Sphenacodontidae (includes the famous Dimetrodon, pictured above from Miller & Harley (2001)) and herbivorous Caseidae, the latter of which were among the largest (3m) tetrapods of their time (Lower Permian) and inhabited dry, upland areas. The sail was composed of the grossly extended spinous processes of the vertebrae and covered by skin, for possible use in thermoregulation, camouflage, or social and sexual communication.

Most “pelycosaurs” went extinct by the Lower Permian after a dominant yet mediocre time on Earth: despite being the very first megaherbivores and megacarnivores, they only managed to spread through western Laurasia (= North America, Europe); by the Late Permian, they had definitively died. One synapsid group subsequently radiated, the therapsids, first known from the Texas Lower Permian Tetraceratops (Amson & Laurin, 2011).

Therapsids quickly rose to prominence and became the dominant organisms until the end-Permian extinction event nearly destroyed them. The three most important groups were:

  • Anomodontia, herbivores and most successful of the ancient therapsids through the diverse and long-lived dicynodonts who had a turtle-like keratinous beak that gave them space for superior mastication abilities and led to their dominance as herbivores;
  • Gorgonopsia, Late Permian carnivores with small eyes and a big and powerful bite; and
  • Cynodontia, the group that includes the mammals.

The aftermath of the end-Permian extinction left only the carnivorous cynodonts and the specialised and dominant herbivorous dicynodonts. The cynodonts radiated in the post-Permian Triassic landscape and by the Late Triassic (225 Ma), the mammaliaforms emerged. These are “mammal-like” insectivores that went on to radiate under the shadow of the dinosaurs in the Jurassic and Cretaceous. They’re called mammal-like because a lot of typically mammalian characteristics are evident in them, especially when it came to cranial and dental features (most important being the dentary-squamosal joint). The KT Event resulted in all but three mammaliaform groups to go extinct. The demise of the non-avian dinosaurs allowed these three groups to radiate and so the modern mammalian fauna was set up, with the monotremes, metatherians (marsupials), and eutherians (placentals) becoming dominant.

I deliberately presented synapsid evolution as a gradual progression towards mammals in this post as summarised in the diagram above (Alberts & Pickler, 2012), despite this being a decidedly bad strategy for teaching the evolution of any taxon. But the evolution of synapsids ending up in mammals is one of the most clear-cut cases of gradual evolution, and this should be stressed. The characteristics known as mammalian characteristics didn’t all pop up in the Tertiary radiation, nor did they suddenly appear in the mammaliaforms, nor in the cynodonts or therapsids.

Instead, they appeared piece by piece, the mammal mosaic becoming gradually more complete at every radiation along the way (Sidor & Hopson, 1998). The most important of these are listed below, in [square brackets] I put in the approximate grade when the changes became noticeable (as far as I’m aware, more knowledgeable readers should correct me!):

  • Changes in skull structure to accommodate larger jaw musculature; included in this is the very important formation of the dentary-squamosal joint, laying the foundation for the ear; [Therapsid]
  • Enlargement of the dentary bone (in the jaw); [“Pelycosaur”]
  • Specialisation of teeth, forming different tooth types; [“Pelycosaur”]
  • Upright stance, as opposed to the crocodile-like sprawl of the other amniotes; [Therapsid]
  • Reduction in ribs to support a diaphragm and thus more active breathing;
  • Secondary hard palate, why we can eat and breathe at the same time; [Therapsid]
  • Respiratory tubinals in the nasal cavity, for accomodating increased metabolic rates by providing an extra pathway for heat release; [Therapsid]
  • Rapid, sustained growth, as opposed to seasonal arresting of growth.

Finally, I want to bring up an enormous pet peeve of mine. You will often hear of synapsids being referred to as “mammal-like reptiles”. This is nonsense, because Reptilia is either defined as synonymous with Sauropsida (with turtles) or with Archosauria (without turtles). Synapsids have nothing to do with reptiles, besides from the fact that they both come from a last common ancestor that was an amniote. Calling them “mammal-like reptiles” is the same as saying that primates are cetaceans. No, they’re not. Cetaceans and primates are their own lines of evolution, and they come from a last common ancestor that was a mammal. Unfortunately, it’s a notation that is traditional and fairly ingrained even in the literature.


Alberts JR & Pickler RH. 2012. Evolution and Development of Dual Ingestion Systems in Mammals: Notes on a New Thesis and Its Clinical Implications. International Journal of Pediatrics 2012.

Amson E & Laurin M. 2011. On the Affinities of Tetraceratops insignis, an Early Permian Synapsid. Acta Palaeontologica Polonica 56, 301-312.

Benton MJ. 2004. Vertebrate Palaeontology.

Miller SA & Harley JB. 2001. Zoology, 5th ed..

Sidor CA & Hopson JA. 1998. Ghost lineages and “mammalness”: assessing the temporal pattern of character acquisition in the Synapsida. Paleobiology 24, 254-273.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.