In my dissertation, I use a comparative approach to exploit an outstanding natural experiment, a suture zone in the rainforest of northeast Australia, to answer questions about speciation and hybridization. The suture zone consists of twenty identified contacts between phylogeographic lineages, mostly within morphologically defined species. Although the contacts in the zone likely formed concurrently in response to Holocene expansion from glacial refugia, the lineage-pairs meeting in the contacts exhibit a wide range of genetic divergences. This natural variation enables analysis of the outcomes of secondary contact at different stages of the divergence process. Importantly, although most studies of speciation focus on lineages that show marked phenotypic divergence, I focus on morphologically cryptic lineages, which, though common in nature, have been understudied in this regard. Through my dissertation, I consider contact zones between six lineage-pairs within four morphologically-defined skink species, Carlia rubrigularis, Lampropholis coggeri, Saproscincus basiliscus, and S. lewisi.

Through this work, I find support for the reality of cryptic species and argue that the presence of cryptic species can suggest a wider plurality of speciation models than we typically consider (Chapter 1). Indeed, by combining multilocus methods and dense sampling, I find that reproductive isolation between phylogeographic lineages scales tightly with divergence (Chapter 2). These results support the widespread, yet previously unsubstantiated, notion that phylogeographic structure of increasing depth represents a continuum towards complete speciation, even in the absence of overt ecologically-driven divergent selection. I extend these results by looking at introgression across the genome, finding that genome-wide selection, driven by selection against hybrids, structures introgression patterns much more strongly than locus-specific selection histories (Chapter 3). By analyzing the sole lineage-pair in this system that exhibits genealogical discordance, I suggest that geographic stability across time is key to driving divergence (Chapter 4). Further, through combining a fine-scale investigation of a single contact zone with simulations and a meta-analysis, I argue that selection against hybrids, in the form of intrinsic genetic incompatibilities, maintains species boundaries at these contact zones (Chapter 5). Finally, as my work is enabled by emerging genomic technologies for non-model organisms, I summarize my genomics approach, and its associated benefits and challenges, as applied to transcriptome data from these lineages (Chapter 6).

In face of environmental changes, species, either move, changing their distributions, or maintain their ranges. Staying in place and facing the new conditions demand adjustments by changes in phenology, acclimation, via behavioral or phenotypic plasticity, or adaptation. Failure to adjust or move consequently leads to population decline and eventual extinction.

The overarching goal of this work is to investigate the mechanisms by lizard species of two tropical genera, Leposom (Gymnophthalmidae) and Enyalius (Leiosauridae), respond to environmental changes at multiple spatial and temporal scales, in the Brazilian Atlantic Forest. I apply a highly integrative approach that combines species distribution modeling, climate data, phylogeography, morphology and thermal physiology and show (1) how geographic isolation, shifts in habitat use, and historical climatic stability explain the patterns of concordance/discordance between genetic divergence and phenotypic disparity in Leposoma scincoides, (2) how climatic variation and current environmental gradients, both at the macro- and microclimatic scales, affect thermal sensitivity across the two species' ranges and (3) how habitat suitability shape species' distribution ranges and spatial patterns of genetic structure in Enyalius catenatus and E. pictus .

Despite deep phylogeographic structure in L. scincoides, and therefore ample opportunity for intra-specific lineages to diverge phenotypically, there is an overall pattern of phenotypic stasis in morphology and physiology. This is most likely due to similar environmental conditions acting on morphology at the climatically stable core of the distribution, and to constraints on thermal tolerances. However, geographically isolated lineages inhabiting more seasonal areas, that were unstable climatically during the Pleistocene, are morphologically distinct. Among those, similarity in morphology was observed between lineages that further broadened their habitat use, occupying more open areas as well as forested habitats. Moreover, morphological variation correlates with climate. These results highlight how distinct phenotypic traits likely to respond to environmental change show very different patterns of variation, conditional on evolutionary lability, plasticity and (possibly) interaction among traits.

Macroclimatic data are not a good proxy for the environmental conditions experienced by the two lizard species in their microhabitats; there is great overlap in temperature availability in similar microhabitats along a macroclimatic gradient in the Central Brazilian Atlantic Forest and climate variables correlated with thermal physiological traits only at the microclimate scale. Both species show narrow thermal tolerances and differences in thermal physiology between the two species can explain some of their ecological differences. Thermal sensitivity is conserved among lineages within L. scincoides, whereas plasticity and acclimation are prominent in E. catenatus' thermal physiology, but the capacity to acclimate to extreme temperatures in the later is limited. These results are alarming if confronted with the rapid anthropogenic changes in habitat cover and climate change predicted to afflict this region.

There is strong phylogeographic structure and high levels of genetic diversity in Enyalius catenatus and E. pictus. These results are consistent with the notion that in this region, Quaternary forest refuges provided stable climatic conditions that promoted persistence of phylogeographic diversity and stable populations. Nevertheless, divergence times, between E. catenatus and E. pictus and among intraspecific lineages, earlier than the late Pleistocene point to a long and complex history of diversification in Enyalius, that requires a variety of factors to explain spatial patterns of genetic structure. Instances of discordance between the species distribution models and genetic structure were pondered carefully in the context of ecological and natural history and the limitations of the models.

My findings show that patterns and mechanisms of divergence and disparity are complex and multifaceted in this region, with geographic as well as ecological components playing important. This study presents an integrated approach to test evolutionary hypotheses that combine phylogeographic and phenotypic data and put it in the context of the landscape, taking into account geographic, environmental, and historical aspects of the habitat. It adds to the growing body of literature that explores phylogeographic patterns of prevalent elements of the Brazilian Atlantic Forest fauna and if/how they responded to climatic oscillations in the Pleistocene. More broadly, it contributes to the growing literature on South America's biogeography and the origin and maintenance of Neotropical diversity.

Background

The glaciation cycles that occurred throughout the Pleistocene in western North America caused frequent shifts in species’ ranges with important implications for models of species divergence. For example, long periods of allopatry during species’ range contractions allowed for the accumulation of differences between separated populations promoting lineage divergence. In contrast, range expansions during interglacial periods may have had homogenizing effects via increased gene flow following secondary contact. These range dynamics are particularly pronounced in the Sierra Nevada, California, given the complex topography and climatic history of the area, thus providing a natural laboratory to examine evolutionary processes that have led to the diversity patterns observed today.

Results

Here we examined the role of late Pleistocene climate fluctuations on the divergence of the Sierra Nevada endemic Alpine Chipmunk (Tamias alpinus) from its sister taxon, western populations of the Least Chipmunk (T. minimus) from the Great Basin. We used one mitochondrial gene (cytochrome b) and 14 microsatellite loci to examine the evolutionary relationship between these species. Mitochondrial sequence data revealed that T. alpinus and T. minimus populations share mitochondrial haplotypes with no overall geneaological separation, and that diversity at this locus is better explained by geography than by species’ boundaries. In contrast, the microsatellite analysis showed that populations of the same species are more similar to each other than they are to members of the other species. Similarly, a morphological analysis of voucher specimens confirmed known differences in morphological characters between species providing no evidence of recent hybridization. Coalescent analysis of the divergence history indicated a late Pleistocene splitting time (~450 ka) and subsequent, though limited, gene flow between the two lineages.

Conclusions

Our results suggest that the two species are distinct and there is no contemporary introgression along their geographic boundary. The divergence of T. alpinus during this time period provides additional evidence that Pleistocene glacial cycles played an important role in diversification of species in Sierra Nevada and North America in general.

The fungal pathogen Batrachochytrium dendrobatidis, or Bd, has been a major driver of amphibian extinctions world wide. This dissertation investigates the effects of Bd on the mountain yellow legged frogs, Rana muscosa and Rana sierrae. These two species occur in the Sierra Nevada in California, and have under gone dramatic declines in part due to the invasion of Bd. Even before the invasion of Bd, populations of Rana muscosa/sierrae have faced habitat loss and fragmentation from introduced predators and habitat fragmentation. The fist chapter of this dissertation investigates how genetic diversity effects $R_0$, the threshold for invasion of a pathogen. I model the case where a naive population faces multiple disease, as well as a single population faced with invasion of a single disease. I also model the effect of overdominance and genetic diversity on R_0. In my second chapter, I use both models and simulations to investigate how genetic diversity effects the final size of an epidemic, and whether or not populations with higher genetic diversity maintain that diversity after an epidemic. In my third chapter, I present data on a series of experiments on how variation in the external source of Bd, or zoospore pool, affects the growth rate of Bd on its amphibian host, as well as host mortality. In my final chapter, I use microsatellite markers to measure genetic diversity in serveral populations of Rana muscosa and Rana sierrae, and test whether or not these populations have experienced historic bottlenecks.

Plethodontid salamanders of the Ensatina eschscholtzii complex have received special attention from evolutionary biologists because they represent one of the very few examples of a ring species, a case where two reproductively isolated forms are connected by a chain of intergrading populations surrounding a central geographic barrier. Ensatina has become a textbook example of speciation, yet there still remain fundamental gaps in our knowledge of this fascinating system. In this study, consisting of three components, I extend previous work on the Ensatina complex in new directions.

In Chapter 1, I conducted a fine-scale genetic analysis of a hybrid zone between the geographically terminal forms of the ring using Bayesian methods for hybrid identification and classification in combination with mathematical cline analyses. F1s and pure parentals dominated the sample. Cline widths were concordant and narrow with respect to dispersal, but there is cytonuclear discordance, both in terms of introgression and the geographic position of mitochondrial versus nuclear clines. Nearly all hybrids possess mitochondrial DNA from one parental type (klauberi suggesting isolation is asymmetrical. Selection against hybrids is inferred to be strong (~21%), but whether this selection is endogenous (genetically-based) or exogenous (environmentally-based) remains to be tested.

In Chapter 2, I investigated the role of late Quaternary climate change on phylogeographic patterns within the Large-blotched Ensatina (Ensatina eschscholtzii klauberi). Intersecting species distribution models constructed under current climatic conditions, as well as two different historical time periods (Last Glacial Maximum, 21 ka and mid-Holocene, 6 ka), predicted stable refugial areas where the species may have persisted throughout climatic fluctuations. Significant phylogeographic structure exists, but geographic structuring of genetic variation by refugia was not supported. Results suggest that populations in putative refugia have not been isolated for very long, or that gene flow may have masked any earlier periods of divergence in allopatry.

In Chapter 3, I conducted a multilocus phylogeographic analysis of the entire Ensatina eschscholtzii complex to reexamine previous phylogenetic hypotheses based on mitochondrial DNA alone. A concatenation approach was used in addition to newer methods that model the relationship between the species tree and the gene trees embedded within them. The concatenated tree was similar to previous mitochondrial trees, identifying well-supported coastal and inland clades, and recovering oregonensis and platensis as paraphyletic. The concatenated tree was not well resolved at the base. Basal relationships recovered by the species tree were well resolved, but most relationships were not well supported compared to the concatenated tree. Results are generally consistent with previous efforts based on mtDNA, but provide further resolution to the Ensatina phylogeny, while highlighting the difficult nature of inferring species trees from samples of closely related populations that are experiencing gene flow.

To date, exon capture has largely been restricted to species with fully sequenced genomes, which has precluded its application to lineages that lack high quality genomic resources. We developed a novel strategy for designing array-based exon capture in chipmunks (Tamias) based on de novo transcriptome assemblies. We evaluated the performance of our approach across specimens from four chipmunk species.

Results

We selectively targeted 11,975 exons (~4 Mb) on custom capture arrays, and enriched over 99% of the targets in all libraries. The percentage of aligned reads was highly consistent (24.4-29.1%) across all specimens, including in multiplexing up to 20 barcoded individuals on a single array. Base coverage among specimens and within targets in each species library was uniform, and the performance of targets among independent exon captures was highly reproducible. There was no decrease in coverage among chipmunk species, which showed up to 1.5% sequence divergence in coding regions. We did observe a decline in capture performance of a subset of targets designed from a much more divergent ground squirrel genome (30 My), however, over 90% of the targets were also recovered. Final assemblies yielded over ten thousand orthologous loci (~3.6 Mb) with thousands of fixed and polymorphic SNPs among species identified.

Conclusions

Our study demonstrates the potential of a transcriptome-enabled, multiplexed, exon capture method to create thousands of informative markers for population genomic and phylogenetic studies in non-model species across the tree of life.