Thanks to N. Brocklehurst for suggesting I take a look at Oromycter (Fig. 1).
Oromycter dolesorum (Reisz 2005, latest Early Permian, Leonardian) is a basal caseid represented by a few skeletal parts, all superbly preserved. It is more primitive than other caseids in that it lacks leafy tooth serrations, has a longer lacrimal (naris-orbit distance) and a larger number of marginal teeth.
Reisz (2005) noted that caseids were unique among synapsids. (* That’s because they were ACTUALLY closer to millerettids and bolosaurids according to the large reptile tree). He noted that caseids appeared late in the evolution of synapsids, but nested at its base and that little is known about the origins of this group (* yet another reason for a large reptile tree with a larger gamut of included taxa: avoiding such by default nestings!).
Oromycter is considered by Reisz (2005) as the most basal member of the caseids. This seems more than reasonable, but has not been tested yet. Outgroup taxa include the basal Caseasauria, Eothyris and Oedaleops, and beyond them Australothyris, Romeria primes and Concordia, not far from the basalmost reptile, Cephalerpeton.
Caseasauria are closer to these millerttids and join other plant-eating basal reptiles (listed above) in the new Lepidosauromorpha, far from the insect-to-meat-eating synapsida (with allowance for Edaphosaurus and the various herbivores in the Therapsida. This needs to be tested on a genus-based level. Might need to rewrite a textbook or Wiki entry.
As always, I encourage readers to see specimens, make observations and come to your own conclusions. Test. Test. And test again.
Evidence and support in the form of nexus, pdf and jpeg files will be sent to all who request additional data.
Reisz R. 2005. Oromycter, a new caseid from the lower Permian of Oklahoma. Journal of Vertebrate Paleontology 25(4):905-910.