Lineage Diversification of Lizards (Phrynosomatidae) in Southwestern North America: Integrating Genomics and Geology
- Author(s): Gottscho, Andrew David
- Advisor(s): Reeder, Tod W
- Gatesy, John
- et al.
In the first chapter of my dissertation, I provide an ultimate tectonic hypothesis for several well-studied zoogeographic boundaries along the west coast of North America, specifically along the San Andreas Fault system. Reviewing the literature, I demonstrate that four Great Pacific Fracture Zones correspond with spatially concordant phylogeographic breaks for a variety of marine and terrestrial animals. I hypothesize that the four zoogeographic boundaries reviewed here ultimately originated via the same tectonic process (triple junction evolution along the San Andreas Fault system), and I suggest how a comparative phylogeographic approach can be used to test this hypothesis. In the second chapter, I investigate the systematics and species delimitation of fringe-toed lizards of the Uma notata complex in the deserts of southwestern North America. Ten nuclear loci were Sanger sequenced and genome-wide sequence and single-nucleotide polymorphism (SNP) data were collected using restriction-associated DNA (RAD) sequencing. I validated five species-level lineages within the U. notata complex, three of which were previously described as full species, one originally described as a subspecies but later synonymized, and one previously documented yet undescribed species from Mohawk Dunes, Arizona, USA. The results support the hypothesis that Pleistocene glacial cycles promoted allopatric speciation via dispersal across a landscape matrix shaped by older tectonic events, but I also recovered evidence for a vicariant role of the Colorado River during the Pleistocene epoch. In the third chapter, I studied the comparative phylogeography of lizards (Phrynosomatidae) of the Baja California Peninsula, including the genera Callisaurus, Petrosaurus, Urosaurus, and Sceloporus. I collected sequence/SNP data from 228 individual lizards from the peninsula and eight islands using RADseq. The estimated divergence dates across co-distributed clades revealed that for each of the five regions where comparisons are possible, at least two or three episodes of divergence are required to explain the observed patterns. I tested for range expansions or bottlenecks associated with Pleistocene glacial cycles or continent-to-peninsula invasions. The results are consistent with the hypothesis that plate tectonic events established a complex landscape matrix with numerous barriers to dispersal that ultimately facilitated divergence, generating the idiosyncratic patterns observed among co-distributed lineages.