Updated February 23, 2015 with a new image of Mycterosaurus.
Mesenosaurus romeri (Efremov 1938, Reisz and Berman 2001) Late Carboniferous to Early Permian ~300 to ~260 mya was originally considered a varanopseid, like Varanops, but it lies outside the varanopsids and outside the synapsids when tested against a larger list of taxa. Here Mesenosaurus was derived from a sister to Archaeovenator and phylogenetically preceded Milleropsis within the Protodiapsida and Eudibamus and Petrolacosaurus within the Diapsida (sans Lepidosauriformes, which nest elsewhere).
The Mycterosaurus question
The Mycterosaurus in figure 1 was illustrated by Williston in 1915. Bones attributed to Mycterosaurus by Reisz et al. 1996 are shown in figure 2. The tooth shapes are not the same. The depth of the maxilla is not the same. Yet the tooth shapes in the Williston image are not the same as those in Heleosaurus and Mesenosaurus. The Reisz et al. images are more similar.
Sometimes one trusts an illustration…
especially if that’s the only available data. Other times, especially if the illustration is old, the trust is reduced. Howecer, the holotype is the benchmark. Fissure fill specimens, disarticulated as they are, and recent figures are typically more accurate. But how do they relate to the holotype?
If anyone has better data on the holotype of Mycterosaurus,
like a photograph of, please send it and the accuracy of the large reptile tree will be enhanced.
Efremov JA 1938. Some new Permian reptiles of the USSR. Academy of Sciences URSS, C. R., 19: 121-126.
Reisz RR and Berman DS 2001. The skull of Mesenosaurus romeri, a small varanopseid (Synapsida: Eupelycosauria) from the Upper Permian of the Mezen River basin, northern Russia. Annals of the Carnegie Museum 70: 113-132. online pdf
Reisz RR, Wilson H and Scott D 1996. Varanopseid synapsid skeletal elements form Richards Spur, a Lower Permian fissure fill near Ft. Sill, Oklahoma. Oklahoma Geology Notes 56 (3):160-170.