Developmental Osteology of the Parafrontal Bones of the Sphaerodactylidae

Well-resolved phylogenetic hypotheses and ontogenetic data are often necessary for investigating the evolution of structural novelty. The Sphaerodactylidae comprises 12 genera of predominantly miniaturized geckos. The genera Aristelliger and Teratoscincus are exceptions, with taxa reaching snout-to-vent lengths far exceeding those of other sphaerodactylids. These two genera possess enigmatic, supraorbital ossifications-parafrontal bones-which are encountered nowhere else among squamates. At the time of their discovery, these structures were believed to be the result of evolutionary convergence. Although relationships between other sphaerodactylids remain unresolved, recent molecular and morphological data have supported a close relationship between Aristelliger and Teratoscincus. We investigated the ontogeny of parafrontal bones to better understand relationships between sphaerodactylid body size and the presence of parafrontals, and to evaluate whether ontogenetic data support the homology of parafrontals between Aristelliger and Teratoscincus. We hypothesize that the parafrontals of Aristelliger and Teratoscincus are homologous and that there is a threshold body size in sphaerodactylids below which parafrontals do not develop, thus explaining their absence from the miniaturized taxa. The presence of parafrontals was investigated in all sphaerodactylid genera using cleared-and-stained, radiographed, and skeletonized specimens. Total surface area of parafrontals was measured for seven species of Aristelliger and six species of Teratoscincus throughout their ontogeny. Histology was used to investigate the cellular composition of the parafrontals throughout their ontogeny. Our data suggest that parafrontals have evolved in parallel from a homologous, parafrontal precursor and that the onset of parafrontal development is not strictly dependent on a threshold body size. Anat Rec, 301:581-606, 2018. © 2017 Wiley Periodicals, Inc.