Mickey Mortimer was kind enough to provide some interesting literature refs regarding the non-reptile status of Diadectomorphs (= Diadectes, Limnoscelis and Tseajaia). That means these taxa would have developed from tadpoles and non-amniote eggs laid in water. The large reptile tree found these taxa to be related to one another deep within the Reptilia, and also related to procolophonids, pareiasaurs, chroniosuchids and turtles. That means diadectomorphs would have laid eggs in dry areas without the young developing form tadpoles. M. Mortimer’s comments followed an earlier post linking caseasaurs and diadetids (but not as sisters, only cousins).
From M. Mortimer’s notes:
“Laurin and Reisz (1995) list several characters of Amniota. Diadectomorphs lack: [my comments follow in brackets]:
– Frontal contacting orbit. [also lacking in Eothyris, Orobates, pareiasaurs and turtles! And… among non-amniotes the frontal contacts the orbit in Cacops and Doleserpeton.]
– Occipital condyle almost as high as it is broad. [Procolophon and turtles also lack this trait.]
– Labyrinthodont infolding of enamel absent (Kemp, 2003). [Diadectes infolding of enamel also absent (double negative here), so this may be a convergent trait with Limnoscelis and labyrinthodonts].
– Axial centrum tilted anterodorsally. [I can’t comment on this hard-to-see trait].
– Cleithrum restricted to anterior edge of scapulocoracoid. [I never knew it wrapped around or did anything else.]
– Presence of three scapulocoracoid ossifications.”[often in basal reptiles (Paleothyris, Eocaptorhinus, the scapulocoracoid is completely fused.]
According to M. Mortimer: “More problematic are-
– Occipital flange of squamosal gently convex. Even if I understood it, they say diadectomorphs’ condition may be an autapomorphy. [such gentle convexity, wherever present, is difficult to determine from drawings and photos].
– Transverse flange bearing a row of large teeth on its posterior edge. Also in Limnoscelis, so could work with synapsid diadectomorphs. [So, Limnoscelis gets a free pass. Captorhinids, caseids, Belebey, Milleretta, etc. etc. also lack a row of large teeth on the transverse flange.]
– Presence of astragalus. They followed Rieppel’s (1993) model, but O’Keefe et al. (2006) showed amniotes have the same astragalar homology as Diadectes. [Okay].
M. Mortimer further reported, “Not to mention diadectomorphs aren’t even sister to caseasaurs in your tree. You have derived diadectids by procolophonids, basal diadectomorphs strewn with Solenodonsaurus, Tetraceratops and chroniosuchians, and caseasaurs sister to various taxa usually placed in Parareptilia, and these three clades are successively less closely related to lepidosaurs.”
No, they’re not sisters (lots of branching in the Millerettids + Caseasauria), but shared a recent common ancestor, Romeria primus.
M. Mortimer also noted that “Berman et al. (1992) and Berman (2000) both suggested diadectomorphs were the sister group of synapsids as opposed to the sister group of amniotes. Kemp (2003) analysed their evidence, concluding-
– A posterolateral corner of the skull table formed entirely or nearly entirely by the supratemporal is only found is Tseajaia, which I note would make it ambiguous synapomorphy is [if] diadectomorphs and synapsids were sister taxa.
– A long posterior expansion of postorbital that contacts supratemporal to exclude the parietal lappet from contacting the squamosal is primitively present in cotylosaurs, including basal sauropsids.
– Presence of an otic trough of the opisthotic is shared, but not by varanopids, which I note isn’t a problem if caseasaurs are basal.
– A deep, nonsculptured component of the tabular which contacts the distal end of a ventrally displaced, laterally directed paroccipital process, enclosing laterally a small, ventrally displaced, posttemporal fenestra is absent in Limnoscelis and Desmatodon, so is more likely convergent in synapsids and Diadectes.“
Nicely put. All I can say is the several hundred traits used to determine the tree topology of the large reptile nested the casesaurs and diadectids where they did because they and 320+ other taxa provided the opportunity and motive. As an experiment, moving all three caseasaurs to the synapsids adds 24 steps. Caseasaurs become strange bedfellows with the basal synapsids and basal protodiapsids derived from a sister to Protorothyris.
The other thing that could be happening here, is this: the Diadectomorpha are on the primitive side of the Amniota and the Synapsida were the first clade to break off from the rest. That alone could indicate a relationship in a traditional sense. Then again, if Berman et al. 1992 and Berman 2000 included casesaurs within their suprageneric Synapsida, then they recovered what the large reptile tree recovered, sans all the other basal reptiles included in the large reptile tree but not included in their analysis.
Thank you for your thoughts, Mickey.
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.