If Azendohsaurus is not a Sauropodomorph Then What is it?

Would you believe an basal archosauromorph?

Flynn, J. J., Nesbitt, S. J., Parrish, J. M., Ranivoharimanana, L., and A. R. Wyss. 2010. A new species of Azendohsaurus (Diapsida: Archosauromorpha) from the Triassic Isalo Group of southwestern Madagascar: cranium and mandible. Palaeontology 53:669-688.

As if the Triassic couldn't get any weirder.  If this discovery does not finally demonstrate the peril of assigning isolated jaw fragments and teeth to various dinosaurian subgroups, I do not know what will.  The placement of Azendohsaurus as a basal archosauromorph demonstrates that herbivory has evolved independently numerous times within Archosauromorpha and was actually much more common in this clade than previously believed.  Some of the primitive cranial features found in Azendohsaurus include a pineal opening, an incomplete lower temporal bar, and palatal teeth.  One unique feature of Azendohsaurus is that the palatal teeth are actually leaf-shaped with denticles, very similar to the marginal teeth.

Skull reconstruction of Azendohsaurus. From Flynn et al. 2010.

Abstract - Here, we describe a new species of Azendohsaurus from the Middle–Late Triassic of Madagascar, extending the geographical range of a taxon known otherwise only by a single species from Morocco. Although Azendohsaurus has consistently been regarded as an early dinosaur (based on various advanced dental and gnathic features resembling those characterizing certain dinosaur subgroups), the relatively complete skeletal material, now available from Madagascar, argues strongly against its dinosaurian affinities. Rather, the retention of numerous primitive cranial and postcranial features indicates a surprisingly early divergence of Azendohsaurus within Archosauromorpha and an unusual mosaic of characters in this taxon. Features considered diagnostic of Sauropodomorpha thus are inferred to occur homoplastically in at least one clade of nondinosaurian archosauromorphs, indicating a complex evolution and distribution of features traditionally thought to be derived within archosaurs. Azendohsaurus has teeth resembling those of both early sauropodomorph and ornithischian dinosaurs, yet also possesses numerous inarguable basal archosauromorph cranial and postcranial attributes. This highlights the risk of uncritically referring isolated, Middle–Late Triassic (or even later), ‘leafshaped’ teeth with denticles to the Dinosauria. Similarly, the occurrence of such teeth in an early diverging archosauromorph indicates that specializations for herbivory originated more frequently within this clade than conventionally assumed. For example, Azendohsaurus and numerous basal sauropodomorph dinosaur taxa share an array of convergently acquired features associated with herbivory, including tooth denticles, expanded tooth crowns, a downturned dentary and the articular located at the ventral margin of the mandible. Some of these features (denticles, expanded crowns and the ventrally deflected articular) are even more widespread among archosauromorphs, including aetosaurs, silesaurs and ornithischian dinosaurs. A downturned dentary also occurs in Trilophosaurus, a taxon further marked by unique specializations for herbivory, including transversely lophate, tricuspid teeth. An array of features associated with herbivory also occur in rhynchosaurs and certain crocodilians (e.g. Simosuchus). This distribution suggests that craniodental features associated with herbivory were much more pervasive across the archosauromorph clade than previously recognized, possibly evolving at least six to eight times independently.