Wikipedia documents a family tree of “the Diapsida” recovered by Modesto and Sues (2004). This is fairly conventional, but based on too few taxa, as you’ll see.
What’s interesting about this tree is if you take the pinks only you have a very good tree of the new Archosauriformes within the new Archosauromorpha that more or less matches the large reptile tree. And the same goes for the light green areas representing the new Lepidosauromorpha. Unfortunately, in the very center of things here you find the suprageneric taxon, “Squamata” and no sphenodontids higher than Gephyrosaurus. If the basal squamate Huehuecuetzpalli was added that might have linked that clade to Drepanosauridae + Tanystropheidae, as in the large reptile tree. If higher sphenodontids were added, that would link them to Trilophosaurus and rhychosaurs, as in the large reptile tree, as we saw yesterday with Lazarussuchus.
There is convergence here. That’s why there’s confusion.
Certainly protorosaurs did converge with tanystropheids, but by adding many more reptile taxa this convergence is revealed. The same goes for nesting lepidosaurs with choristoderes.
More taxa equals greater resolution and fewer strange bedfellows. Try it and see. It’s a simple experiment that any amateur or professional can do. If I can do it, you can do it.
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.
Modesto SP and Sues HD 2004. The skull of the Early Triassic archosauromorph reptile Prolacerta broomi and its phylogenetic significance. Zoological Journal of the Linnean Society 140 (3): 335.