Studies on the phylogenetic relationships of snakes and lizards are plagued by problematic characterizations of anatomy that are then used to define characters and states in taxon-character matrices. State assignments and character descriptions must be clear characterizations of observable anatomy and topological relationships if homologies are to be hypothesized. A supposed homology among snakes, not observed in lizards, is the presence of a crista circumfenestralis (CCF), a system of bony crests surrounding the fenestra ovalis and lateral aperture of the recessus scalae tympani. We note that there are some fossil and extant snakes that lack a CCF, and some extant lizards that possess a morphological equivalent. The phylogenetically important upper Cretaceous fossil snake Dinilysia patagonica has been interpreted by different authors as either having or lacking a CCF. These conflicting results for Dinilysia were tested by re-examining the morphology of the otic region in a large sample of snakes and lizards. An unambiguous criterion arising from the test of topology is used to define the presence of a CCF: the enclosure of the ventral margin of the juxtastapedial recess by flanges of the otoccipital (crista tuberalis and crista interfenestralis) that extend forward to contact the posterior margin of the prootic. According to this criterion D. patagonica does not possess a CCF, therefore, this anatomical feature must have arisen later during the evolution of snakes. J. Morphol. 275:1187–1200, 2014. © 2014 Wiley Periodicals, Inc.