Working with and looking at various palates close to that of Anningasaura, Endennasaurus and Askeptosaurus, brought to light several interesting variations on the enaliosaur theme, including the solidification of the palate in nothosaurs and the disassociation of the ectopterygoid from the pterygoid in Miodentosaurus.
This is the ancestral enaliosaur recovered by the large reptile tree. It has a basic diapsid palate, comparable to those in Petrolacosaurus and Araeoscelis with all the elements in the standard, plesiomorphic positions.
Here represented by Pachypleurosaurus, Lariosaurus, Simosaurus, the nothosaurs share some palate configurations. The pterygoids grow together medially causing the basisphenoid to disappear in palatal view. Note that Pachylpleurosaurus alone does not have a maxilla, naris contact, nor does it have a maxillary palate shelf, nor does it have a ectopterygoid that contacts the maxilla and jugal. These may be errors. The mandible covers the rim of the palate in all examples of Carroll 1985. I divided the palate in half in figure 1 with the left half reflecting one possibility and the right side reflecting Carroll’s interpretation.
Anningasaura and Pistosaurus
Like Claudiosaurus, the braincase remained visible in Anningasaura and Pistosaurus These two taxa share a similar pterygoid with posterior processes. Both share a narrow rostrum and broader cheeks.
Like Claudiosaurus (and the alternative interpretation of Pachypleurosaurus) the suborbital fenestra was retained in the palate of Plesiosaurus.
The odd, turtle like Sinosaurophargis was a sister to the Placodontia but derived from Claudiosaurus. The skull was broad with a narrow snout. The ectopterygoid enlarged and the palatine reduced.
Derived from Claudiosaurus, the rostrum elongated in Mesosaurus. The ectopterygoid and palatine were fused and both were anterior to the pterygoid. The pterygoid had transverse processes. The maxilla had a long contact with the vomers, shifting the internal nares posteriorly.
The palate of Wumengosaurus was similar to that of Mesosaurus, but with an unfused palatine and ectopterygoid. The ectopterygoid extended further anteriorly than the palatine, extending laterally to the internal nares. The pterygoid transverse processes angled anteriorly.
Hupehsuchus and Ichthyosaurs
Descending from a sister to Wumengosaurus, Hupehsuchus and ichthyosaurs both shifted the palatines anteriorly while reducing them. The ectopterygoid filled the vacancy.
Descending from a primitive mesosaur/claudiosaur, the palate of Askeptosaurus had tiny, anteriorly displaced internal nares (like mesosaurs) and suborbital fenestra (like Claudiosaurus). The premaxilla was long, as in mesosaurs. The palatines were reduced yet contacted the maxilla, as in Claudiosaurus. More detail here.
Toothless Endennasaurus is difficult to restore due to the presence of robust mandibles covering the palate rims and a big break that crosses the middle. Here the darker colors represent restored areas. The pterygoids were dorsal to the conjoined palatines. The pterygoid posterior processes were elongated and robust. The ectopterygoids apparently lose contact with the pterygoids and are placed between the palatines and jaw rims, or are absent all together. Then again, a connection might remain. That break prevents knowing. If present, the lateral portion of the ectopterygoids cannot be seen on the fossil due to the overlying mandibles. More detail here.
In Miodentosaurus the ectopterygoid loose all connection to the pterygoids and are placed between the palatines and the jaw rim, as in Endennasaurus by convergence.
In Clarazia the Y-shaped palatines are set between the pterygoids and premaxilla/maxilla shelf with its double row of low, round crushing teeth.
The variety of palates in this clade is interesting. Obviously they reflect different diets, as do the teeth. As noted earlier, the similarities in the Mesosaurus and ichthyosaur palates are notable.
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.