Updated February 17, 2015 with new image of Odontochelys
After a weekend reconstructing Odontochelys as a living animal (Fig. 1), rather than the prior “roadkill” I thought it prudent to take another look at this key taxon in the origin of turtles.
There may be other solutions here.
Unfortunately the torso of Milleretta RC70 is not known. Like the smaller RC14 specimen, it too may have had expanded costal ribs.
(Note: At this point I started working on Stappenbeck’s Stephanospondylus and discovered, among many other things, large costal plates on the ribs that Stappenbeck identified as clavicles.)
Figure 1. Odontochelys with a newly reconstructed skull.
In any case,
further differences between Odontochelys and Proganochelys (Fig. 2) became identifiable such as the tiny clavicles (Figr. 1 in blue), the small ilium (in green), the lack of a pedal digit 5 and a more parsimonious configuration of the dorsal ribs. Note the hind limbs were not enclosed within the expanded torso proto-shell.
Figure 2. Proganochelys and Proterochersis, two Traissic turtles, both to scale.
It’s also important to realize
that Proganochelys, despite its antiquity, is really a derived turtle in the lineage of the giant horned turtles with club tails like Meiolania, (Fig. 3) a lineage that persisted until rather recent times. More plesiomorphic turtles, like the slightly older Triassic Proterochersis Fig. 2) demonstrate that more “common”-looking turtles and likely a wide variety of turtles were already present then and earlier. We just haven’t found them yet.
Figure 3. Meiolania is another club-tailed, short-toed turtle like Proganochelys.
References
Joyce WG, Schoch RR and Lyson TR 2013. The girdles of the oldest fossil turtle, Proterochersis robusta, and the age of the turtle crown. BMC Evolutionary Biology 13: 266. doi:10.1186/1471-2148-13-266 abstract
The Pterosaur Heresies 