Date of Degree


Document Type


Degree Name





Frank T. Burbrink

Committee Members

Michael Hickerson

Ana Carnival

Elizabeth Alter

Evon Hekkala

Chris Raxworthy

Subject Categories

Biodiversity | Bioinformatics | Computational Biology | Desert Ecology | Evolution | Genomics | Integrative Biology


ecological speciation, comparative phylogeography, statistical phylogeography, community ecology, squamate


Understanding the process of speciation is of central interest to evolutionary biologists. Speciation can be studied using a phylogeographic approach, by identifying regions that promote lineage divergence, addressing whether speciation has occurred with gene flow, and when extended to multiple taxa, addressing if the same patterns of speciation are shared across codistributed groups with different ecologies. Here I examine the comparative phylogeographic histories and population genomics of thirteen snake taxa that are widely distributed and co-occur across the arid southwest of North America. I first quantify the degree to which these species groups have a shared history of population divergence across a well-documented phylogeographic barrier, the Cochise Filter Barrier (CFB), and estimate the timing of divergence for each taxon pair using a single locus. This study reveals 1) substantial population structure in these snake groups, 2) climate explains the greatest amount of genetic divergence, and 3) that the CFB has likely been important in lineage formation and species diversification. Although these species groups broadly share population structure, multiple methods of divergence-time dating illustrate that there is strong support for asynchronous diversification and little concordance among timing estimates.

In order to address how speciation has occurred within these species groups, I generated a sub-genomic dataset using a reduced representation approach, genotyping-by-sequencing. I then used coalescent population genetic techniques to model historical demographic patterns across these thirteen species groups for 5,496 – 21,259 unlinked single nucleotide polymorphisms (SNPs), assessing congruence in spatial genetic structure, timing of divergence, and the mode of speciation across the desert southwest of North America. I first assessed population structure and tested for a signature of isolation-by-distance, then tested the validity of these groups as distinct species using a coalescent method that accommodates the species phylogeny as well as incomplete lineage sorting. Tests of selection, using a method that estimates residual levels of population structure, were also implemented because selection can confound demographic inference. I then used demographic model selection to examine mode of species divergence in each taxon, assessing alternative models that included (a) strict isolation, (b) divergence with secondary contact, and (c) divergence with continuous gene flow. Results from this study suggest that all populations identified via clustering methods represent distinct species with non-concordant contact zones across each species group. In addition, there is significant isolation-by-distance in nearly all species groups. Importantly, model selection demonstrates with strong support that speciation occurred with continuous gene flow, suggesting parapatric speciation is common across this assemblage of snake species. Therefore, in contrast to the classic allopatric model of speciation, which has been the paradigm mode in vertebrates, I demonstrate that speciation with gene flow may be common at well-documented phylogoegraphic barriers. Additionally, diversification due to large-scale climatic oscillations during the Quarternary does not explain the pattern of speciation within this region.

Parapatric speciation is often considered to be the result of divergent natural selection across a heterogeneous landscape. However, isolation-by-distance has been shown to produce patterns of parapatric speciation without ecological gradients, geographic barriers, or disruptive natural selection. Within this system, I found strong support for parapartic speciation across a heterogeneous landscape, yet with a significant signal of isolation-by-distance. Therefore, I also tested whether speciation across the arid southwest was mediated by divergent ecological speciation or occurred neutrally, simply due to geographic distance separating populations. These two processes propose different, testable predictions: ecological speciation posits that reproductive isolation positively correlates with adaptive ecological divergence and divergence time, while neutral processes imply reproductive isolation will be positively correlated with divergence time and not ecological divergence. I test these alternative hypotheses using metrics of reproductive isolation from the previously generated population genomic data. Ecological divergence was estimated using ecological niche models and morphometrics defined by the degree of divergence in vertebral number and head shape. Both multiple linear regression models and Bayesian model averaging were used to assess how reproductive isolation scales with divergence time and ecological divergence. I find that reproductive isolation co-varies with head morphology across these thirteen sister species pairs, such that as the strength of reproductive isolation increases, divergence in head shape also increases. This work provides a potential mechanism for parapatric speciation and suggests divergent natural selection on ecology is important in driving lineage formation and speciation.

In summary, the results from these analyses advance our knowledge of community diversification by illustrating how taxon groups have independent histories (i.e. asynchronous divergence times and independent primary contact zones) across a region notable for influencing patterns of speciation. These differences are likely a result of species-specific ecological preferences and physiological tolerances. Despite this, there is a shared, yet seemingly uncommon geographic mode of speciation: parapatry. Finally, this shared mode is likely mediated by a common mechanism of divergent natural selection, which can be detected in differences in head shape morphology. This suggests that parapatric speciation is being driven by adaptation to different environments across the arid southwest of North America.


To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.