As I mentioned in the post about the dire toad, brachiodonts evolved in two clades of reptiles and one clade of mammals called Plesiotheroidea. Plesiotheroidea is named for Plesiotherium, so-called because it means “almost mammal,” and also because it greatly resembles a plesiosaur. Colloquially, the animal is known as a burrowing walrus, because its brachidonts resemble tusks from a distance, it has a face like a seal, and it likes to poke its head out of holes in the ice and then disappear. However, the name “burrowing walrus” is misleading on both counts, as the animal doesn’t make the holes, and instead pokes its head out of existing holes to breathe and snatch prey, and it isn’t a walrus, not by a longshot. Before I can explain exactly what it is, I have to give you a brief lesson on mammalian evolution.
The above cladogram is a screenshot from the Wikipedia entry on Synapsids. As you can see, there are quite a few intermediate steps between basal synapsids and mammals. The most famous transitional form from this group, the sail-backed sphenocodont Dimetrodon, isn’t even pictured.
On Varanganska as on Earth, amniotes began as anapsids, possessing a skull with no unnecessary holes in it. For various reasons, some descendants developed holes in the temple called temporal fenestrae. Those with one pair of extra holes, the synapsids, eventually gave rise to mammals, while those with two extra pairs of holes, the diapsids, gave rise to true reptiles, including lizards, crocodiles, and dinosaurs. Early examples of each, however, looked and acted much like lizards and weren’t all that different fundamentally. Advanced synapsids, however, developed many new characteristics that separate them from the classic idea of what a reptile is. One subset, called therapsids, were endothermic, and one subset of those, the cynodonts, had spines suited to undular flexation, rather than serpentine. This is the point at which evolution on Varanganska no longer matches the evolution on Earth, simply because of the hexapod paradigm. While an extra pair of limbs would make little difference to a salamander or lizard, owing to the serpentine flexation of the body, it causes problems for mammals and non-mammalian cynodonts, so there was immense selective pressure to either specialise or lose one of these pairs of limbs, as certain reptiles eventually did also. Plesiotheroidea represent the result of this exact process, having turned their front limbs into phyrangeal jaws called brachiodonts. According to conventional classification, this would mean that plesiotheres are one of only two groups of mammals to have six limbs, and the only mammals to have highly differentiated limbs, but there is a problem with that: plesiotheres aren’t mammals at all, because they don’t produce milk.
Below is an overly-simplified cladogram showing the relationships of Varanganskan mammals to each other. Without the means to zoom in, there is no way you’re going to be able to read the labels, but you get the idea (I hope).
The only reason that Plesiotherium is even called a mammal is because it has fur and bears live young, but most of its ancestors laid eggs, while contemporary mammals were bearing live young at that time. The reason for this is that, while most fish and amphibians lay eggs, amniotic obligate swimmers must all give live birth. For instance, sea turtles lay eggs on beaches, but icthyosaurs, which were also marine reptiles, had to give live birth because they were obligate swimmers (and some could be even bigger than sperm whales, which can’t even survive being beached the way dolphins can). What this means, in terms of phylogeny, is that Plesiotheroidea split off before the emergence of prototherians, also known as monotremes. Monotremes produce milk, so they are definitely mammals, but they lay eggs. Technically, plesiotheres are non-mammalian cynodonts, just as weasel-rats are, but they are much more distantly related to tritylodonts than to mammals.
A much more closely related group to “typical” Varanganskan mammals is the clade Paratheria. The name is somewhat misleading, meaning “beside mammals,” since these are true mammals by all measures. In fact, they are placental mammals, but evolved as an unusual offshoot, and are more distantly related to the main placental superorders than those superorders are to each other. Unlike all other true mammals on the planet, paratheres have six legs, and the extra structural support means that they can grow much larger than other mammals. Among Paratheria can be found numerous animals showing convergent evolution with much more familiar fauna, particularly ungulates. One example is Optoceras, colloquially known as a glazorog. Both names literally mean “eye-horn,” but in different languages. Glazorogs are what Rossbera has instead of rhinos. It is a stout, three-toed ungulate, and depending on the species, two or three horns. There may or may not be a horn on the nose, but there are always long horns just above the eyes, which grow up at the root, but then curve to that the points are facing forward. Another example is Parahippus panpater, which means “All-Father’s near-horse.” Colloquially, this animal is known as a sleipnus, and looks like a horse with six legs. What? Don’t look at me like that, I can rip off Norse mythology just like anybody else.
As I mentioned spinal flexation earlier, this is the very reason that the sleipnus and glazorog both appear to have doubled forelegs, whereas more “primitive” forms, such as the weasel-rats, have legs that are evenly spaced over the length of the body, just as Varanganskan lizards do (other than snakes, of course). Most the animals in this clade tend to have bulky upper torsos, while being much more slender overall than their tetrapodal analogues. For this reason, while most Rossberan cavalry units employed the stocky dostrop, elite riders would seek out sleipni as much as they could. Sleipni are faster and more agile, but while individual animals may be tamed, the species itself cannot be domesticated. Elite warriors of Skhara, for instance, would embark on long treks into the wilderness to commune with and ultimately win over the loyalty of a sleipnus, which they would then ride back home, where they would be welcomed with a great celebration. To ride a sleipnus was, more or less, required for a Skharan warrior to ever become a general.
I am really bad at drawing mammals, hence my use of rather simplistic templates that someone else created, and that I merely modified. Arthropods are easier, to be perfectly honest, but the fact that drawing software is tedious to use doesn’t help. I do plan on eventually going back and making much better drawings, and the larger cladogram with all of my made-up fauna is still under construction.