Kammerer, C. F., Flynn, J. J., Ranivoharimanana, L., and A. R. Wyss. 2012. Ontogeny in the Malagasy Traversodontid Dadadon isaloi and a Reconsideration of its Phylogenetic Relationships. Fieldiana Life and Earth Sciences 5:112-125. doi: http://www.bioone.org/doi/abs/10.3158/2158-5520-5.1.112
Abstract - New craniodental material of the traversodontid Dadadon isaloi from Middle/Upper Triassic basal “Isalo II” beds of southwestern Madagascar is described. These specimens reveal several new autapomorphies of Dadadon, including paired foramina on the frontal near the anterior border of the postorbital and lower incisors with denticulated distal margins. The new material covers a broad size range, providing the first information on ontogeny in Dadadon. Larger (presumably older) specimens of Dadadon isaloi have more postcanine teeth, relatively longer, narrower snouts, and a higher degree of cranial ornamentation than smaller specimens. Postcanine replacement in Dadadon was similar to that of other traversodontids: new teeth erupted at the posterior end of the postcanine tooth row and moved forward. Using information from the new specimens, the position of Dadadon was tested in a new phylogenetic analysis of traversodontids. In the new analysis, Dadadon is strongly supported as a member of a clade also including the South American taxa Massetognathus and Santacruzodon, here named Massetognathinae subfam. nov. This clade is diagnosed by the presence of denticulated lower incisors, relatively small canines, three cusps in the labial margin of the upper postcanines, and low, flat skulls. Massetognathinae is the sister-group of Gomphodontosuchinae, which includes Gomphodontosuchus, Menadon, Protuberum, Exaeretodon, and Scalenodontoides. The Laurasian traversodontids (Arctotraversodon, Boreogomphodon, and Nanogomphodon) form a clade that is the sister-taxon of Massetognathinae + Gomphodontosuchinae. Denticulated incisors evolved multiple times in traversodontid evolution (in massetognathines and Arctotraversodon), and thus this group represents another possibility (besides various archosauromorphs) to be considered when attempting to identify isolated Triassic teeth with denticulated carinae lacking cingula.