Perhaps the most evident conversion of genomic information into functional, morphological phenotypes in an animal occurs during organogenesis, and the study of vertebrate tooth development provides a phenotypically diverse system for which the mechanisms for patterning and morphogenesis have been extensively studied. An understanding of the developmental basis for evolved differences between teeth in different anatomical and phylogenetic contexts brings complementary information to our knowledge of odontogenic mechanisms. Examining difference, or variation, allows for the validation of hypothesized developmental mechanisms, identification of mechanistic flexibility that could be available to evolution or bioengineering, and the redefinition of phenotypes to better align with the natural biological variation available.

This dissertation examines the development of the dentition in the frog and emerging developmental model Silurana (Xenopus) tropicalis, including the first gene expression data for odontogenesis in any amphibian. Comparative data for the evolution of dental phenotypes are assembled from descriptions of tooth initiation, dentition patterning, and adult craniodental variation phenotypes, addressing developmental questions at population, subfamily, and phylum levels.

Using hematoxylin and eosin-stained histological sections and whole mount preparations of larval S. tropicalis jaws, I demonstrate that individual tooth initiation is broadly similar to that documented for phylogenetic relative Xenopus laevis, but that the process is temporally shifted relative to external developmental traits in the Nieuwkoop and Faber staging system. Furthermore, patterns of tooth initiation in S. tropicalis reveal a lack of synchrony in alternating tooth positions and dynamics that were previously undetected. The frequent presence of ‘twinned’ tooth germs in whole mount preparations argues against a robust model of local inhibition directing tooth initiation in this species. These findings rule out two hypothesized developmental mechanisms for tooth initiation in S. tropicalis that were derived from data in X. laevis and other homodont vertebrates.

In another investigation of first generation tooth development, I examine the expression of Sonic hedgehog (Shh) , a marker for several phases of odontogenesis across vertebrates. I demonstrate the utility of comparing ‘natural experiments’ in development with what is known from more anatomically conservative developmental models. In particular, I use the fact that S. tropicalis teeth do not initiate until just before metamorphosis as a case where tooth formation and mouth formation developmental programs are dissociated from one another to evaluate the current consensus odontogenic model for Shh. With in situ hybridization data from S. tropicalis, I fail to detect a Shh-expressing odontogenic band prior to tooth formation, counter to predictions from the consensus model. A review of published functional data and the correspondence between an odontogenic band domain and the presence of functional teeth in other vertebrates reveal several other taxa for which the consensus model appears insufficient to account for variation in the distribution of the marginal dentition.

Finally, I explore the ability to infer developmental processes from patterns of adult craniodental variation in three S. tropicalis genetic strains raised in captivity. Osteological measurements and tooth counts are analyzed for patterns of covariation at the functional phenotypic levels of the cranium, the jaw, and the tooth–bearing bone. I demonstrate that the bimodal sexual body length dimorphism does not carry over to any cranial metric trait measured; systemic factors affecting cranial length can explain much of the difference between male and female traits, which are distributed unimodally. Patterns of covariation with cranial size, size–adjusted patterns of pairwise phenotypic correlation, and significant differences between genetic strains all suggest a relative independence of variation in the premaxilla and maxilla in S. tropicalis, and I document evidence for a functional jaw module, in which the tooth row and jaw bones correlate when summed across the jaw, but in which tooth and jaw phenotypes lack integration at the level of individual tooth–bearing bones.