Connectivity refers to the amount of gene flow present between populations of the same species. The transfer of genetic material between populations allows species to increase their genetic diversity, thus allowing advantageous mutations to spread. When gene flow between populations becomes restricted, each population may evolve independently, diverging into what could eventually become two different species. In reef fishes, speciation events such as these are common particularly in peripheral populations that disperse to remote islands. Species that occupy small geographic ranges (e.g., endemics) tend to have low genetic diversity, thus being more vulnerable to strong environmental changes. A region where connectivity has been relatively understudied is the Tropical Eastern Pacific (TEP), which exhibits high level of endemism of shore fishes and is biogeographically isolated from other provinces. Despite having had multiple speciation events of marine organisms in the TEP, most connectivity studies of this region show high gene flow across long distances. However, the majority of these studies rely on single mitochondrial DNA markers. In this study I use Restriction Site Associated DNA sequencing (RADseq) to obtain 1000s of loci per individual to compare intra- and inter-specific populations of Holacanthus angelfishes in the TEP. For my first chapter, I assembled the whole genome of Holacanthus passer using a combination of high coverage long- (Oxford Nanopore) and short-read (Illumina) technology to further our understanding of the evolutionary history of H. passer. The draft genome resulted in a final assembly of 583 Mb contained in 476 contigs with a contig N50 length of 5.7 Mb. The genome contained 97.5% complete conserved actinopterygian orthologs, making it comparable, if not superior, to many chromosome-level genome assemblies of fishes. Using whole-genome sequence information, the demographic history of H. passer indicates a population expansion in the TEP preceded the last glacial maximum, as supported by other studies. For my second chapter, I used RADseq markers to detect genetic breaks and genetic diversity hotspots for intra-specific populations of H. passer across the TEP. I obtained a total of 19,809 polymorphic loci that revealed high gene flow along the TEP coastline (FST = 0.00) as predicted by the literature. However, pairwise differentiation detected weak but significant structure between Panama and the Sea of Cortez (0.002 < FST < 0.005; 0.007 < p < 0.043), driven principally by isolation by distance. Interestingly, we detected a temporal discord between individuals collected in Panama 10 years apart and did not show this same genetic signal. In addition, we detected 28 outlier loci that revealed subtle genetic signatures that differentiated populations from the mainland and oceanic islands. My third chapter took a broader approach to assess the inter-specific genomic signatures of Holacanthus angelfishes in the TEP. H. passer is mainly found on the mainland in the TEP, while its sister species H. clarionensis is endemic to the Revillagigedo Archipelago, and H. limbaughi is endemic to Clipperton Island. RADseq markers detected three hybrid individuals between H. passer and H. clarionensis but none with the third sister species, H. limbaughi. Moreover, equal amounts of ancestral variation of H. passer among H. clarionensis individuals and the lack of F2 or back-cross hybrids suggests that hybrids are sterile and provides evidence of incomplete lineage sorting. Although H. limbaughi and H. clarionensis are presumed to have diverged around the same time from H. passer, H. limbaughi’s smaller effective population size may have led to a faster rate of lineage sorting. Overall, this study highlights the power of using genome-wide markers (e.g., RADseq) to deliver a higher resolution perspective on the population dynamics within the TEP while giving insight into the evolutionary mechanisms that drove divergence of Holacanthus species.

Here, in embiotocid surfperches I have investigated diverse evolutionary processes on different genomic scales using advances in massively parallel DNA sequencing technologies. First I reconstructed phylogenetic relationships among all genera and 21 out of 23 embiotocid species using restriction-site associated DNA sequence (RADseq) markers. In Embiotocidae I found that RADseq supermatrices that retained 91% of orthologous markers across sampled species, which corresponded to 523 loci, yielded trees with the highest support values. The resulting phylogenetic hypotheses support a scenario where embiotocids first diverged into clades associated with sandy and reef habitats during the middle Miocene (13–18 million years ago) with subsequent invasions of novel habitats in the reef associated clade, and northern range expansion in the Northwest Pacific. In all cases, radiations occurred within specific habitats, a pattern consistent with niche partitioning. For my second chapter I characterized the complete mitochondrial genome of the black surfperch, Embiotoca jacksoni. The genome contains 13 protein-coding genes, two ribosomal RNA genes, 22 transfer RNA genes, and the non-coding control region (Dloop), the gene order of which is identical to that observed in most vertebrates. I then compared the protein-coding mitochondrial DNA gene sequences of E. jacksoni with two other embiotocid surfperches with available complete mitochondrial genomes, Cymatogaster aggregata and Ditrema temminckii. Across all mitochondrial protein-coding genes in surfperches the weighted average substitution rate was 2.079% per million years and average dN/dS ratios for each protein-coding gene ranged from 0.016 in CytB to 0.608 in ND3. Substitution rates and dN/dS ratios were relatively high for ATP8 compared to other protein-coding genes. Although most protein-coding genes showed signals of purifying selection, I found evidence for positive selection in ND3 in E. jacksoni. Finally for chapter three I performed exploratory genome scans for selection in four species of Embiotocid surfperches between populations in Monterey Bay, California, USA and Punta Banda, Baja California, Mexico again using RADseq markers. These localities are on opposite sides of a well-known biogeographic break and experience significantly different sea surface temperatures (SST) as well heterogeneous species assemblages. Using a FST outlier approach, I detected strong signals of intraspecific balancing and divergent selection as well as clear evidence for interspecific parallel selection with some loci aligning to known proteins. These results suggest that California surfperches represent an ideal system for investigating different forms of selection in a natural marine environment.

While it was once widely accepted that there is plentiful larval dispersal and high levels of connectivity in marine populations, studies detecting self-recruitment and local larval retention in reef fishes have revealed that not all marine organisms demonstrate broad-scale dispersal. Isolated oceanic islands with their abundance of endemic species serve as excellent systems for studying the factors that may be influencing dispersal in marine populations. Furthermore, understanding patterns of dispersal in endemic species is relevant to conservation and management efforts that seek to protect these unique and vulnerable species. This dissertation employed molecular techniques to analyze the phylogeography of damselfishes endemic to the Hawaiian Archipelago and to draw conclusions about patterns of connectivity in endemic reef fishes. The endemic damselfishes that have been surveyed throughout the archipelago thus far all exhibit population structure (Chapter 1). This is consistent with the assumption that the low dispersal ability of endemic species results in population structure. However, some Hawaiian endemic reef fishes lack genetic structure in the archipelago, so population structure may be characteristic of endemics only in certain taxonomic families, such as the Pomacentridae. In contrast to their endemic counterparts, widespread damselfishes that have broad distributions across the Indo-Pacific show high levels of connectivity across the Hawaiian Archipelago (Chapter 2). This side-by-side comparison of widespread and endemic damselfishes lends support to a proposed correlation between range size and dispersal ability, at least within the spatial scale of the archipelago. These widespread damselfishes, together with several species of widespread reef fishes that lacked population structure in the Hawaiian Islands, show genetic differentiation across the spatial scale of the Indo-Pacific. Thus, at a larger spatial scale (between archipelagoes), the relationship between range size and dispersal ability is more tenuous. Patterns of connectivity even vary with spatial scale in the endemic damselfish Chromis verater found on shallow and mesophotic reefs in the Hawaiian Archipelago and Johnston Atoll (Chapter 3). The dispersal abilities of Chromis verater do not appear to limit it in terms of vertical connectivity (7-113 m in this study), but there are some restrictions to horizontal connectivity across the Hawaiian Archipelago (2600 km) and between the archipelago and Johnston Atoll (separated by 860 km). Overall, the results from this dissertation lend support to the assumption that the low dispersal ability of endemic species results in less connectivity and more population structure than in widespread species. However, it remains unclear whether this applies to all reef fish families or if it holds true in locations besides the Hawaiian Archipelago. Also, these results illustrate the importance of considering spatial scale and phylogenetic constraint when drawing conclusions about dispersal in reef fishes.

The formation of the Baja California peninsula separated the distributions of 19 marine fishes into disjunct Pacific and Sea of Cortez populations. Similarly, their Pacific distributions cross phylogeographic points that diminish the genetic connectivity of their populations. This resulted in multiple species experiencing a gradient of gene flow and an extraordinary framework to study mechanisms of divergence and signals of selection under different scenarios of isolation. Genetic isolation in these species has previously been studied using only a handful of markers. In this dissertation, Restriction Site-Associated DNA (RADseq) is used to genotype thousands of genome-wide makers to study the evolutionary history, characterized genomic patterns of divergence among populations, and search for signals of drift and selection, in four of these species: the sargo, Anisotremus davidsonii (ADA); the longjaw mudsucker, Gillichthys mirabilis (GMI), the California sheephead, Semicossyphus pulcher (SPU); and the zebraperch, Kyphosus azureus (KAZ). While, no evidence of isolation was found for SPU, estimated dates the disjunction of ADA, GMI, and KAZ, were 60, 284, and 23 thousand years, respectively. A dispersal episode where species migrated northwards and then became isolated by the warming of the seawater temperate in the south of the Baja California peninsula at the end of the last glaciation period, and a vicariant isolation resulting from the closure of a mid-peninsular seaway, are the plausible historical events that produced the current distributions of these species. Lower than expected levels of genomic gene flow (significant p-values) were seen across Point Conception in GMI (FST=0.15), Punta Eugenia in ADA (FST=0.02), and across the peninsula in KAZ, GMI and ADA (FST=0.03, 0.11, and 0.23, respectively). Furthermore, when comparing disjunct populations, 19 to 46 % of outlier loci matched coding genes in all species and analyses identified 15 genomic regions, potentially involved in processing environmental information, metabolism, immune response, and possibly reproduction, diverging in more than one of these species. Results are interpreted as supporting (1) the idea that ADA and GMI disjunct populations are in the initial phases of allopatric speciation and (2) the presence of convergent selection in these species.

The interactions among organisms shape the earth's biodiversity in striking ways, and these interactions are known to vary substantially over both time and space. Identifying the scale of variation in species interactions allows for a better understanding of the origins and organization of biodiversity, and informs expectations of the dynamics of these systems. In order to study the scale of variation in species interactions, I focused on the anemonefishes, a group of coral reef damselfishes engaged in a mutualism with sea anemones. I used three complementary approaches. First, I used population genetic methods to show that populations of anemonefishes are genetically isolated over relatively small spatial scales in the Mozambique Channel. Second, I employed data from social media to reveal spatial variation in the interactions between anemonefishes and their host sea anemones. Third, I reconstructed the phylogenetic history of the anemonefishes to examine their pattern of diversification in the context of both ecological and geographic processes. The relatively low dispersal potential of anemonefishes likely contributes to both the spatial structure among populations within species, and the strong geographic pattern of diversification among species. In turn, the spatial variation in diversity within the anemonefishes has led to variation in their interactions with sea anemones.

Studies on the ecology and biogeography of seamounts and oceanic islands have advanced substantially in the last 60 years. However, few seamounts have been scientifically characterized, with basic aspects of their biodiversity still unknown and many hypotheses not empirically tested. Consequently, the role of seamounts in the evolution for marine species is still unclear. In the south Atlantic, the Vitória-Trindade Chain (VTC) extends ca. 1,200 km offshore the Brazilian coast. For a long time, most of the biological information available was limited to the emergent islands, where a high richness and endemism is found. The seamount chain, situated between the islands and the mainland, is thought to contribute importantly to the regional biodiversity, and has been suggested as stepping-stones, repositories and museums for coral reef species. The goal of this dissertation is to characterize the biogeographic patterns of reef fishes that inhabit the VTC and test hypotheses to explain those patterns. The results reveal large mesophotic reefs sheltering a very high biodiversity, where new records, depth range extensions and new species were reported. The regional biogeography is very complex; species richness does not decrease with distance from the mainland and it is related instead to seamount and island environmental characteristics and ecological traits of the species. Speciation is not only resulted from the islands isolation, as many endemics are widely distributed along the seamounts, and selection and adaptation are stronger drivers of genomic differentiation than is isolation. In conclusion, oceanic islands would unlikely exhibit the biogeographic patterns we observe today in the absence of seamounts as intermittent stepping-stones, reservoirs of coral reef biodiversity.

Damselfishes (Family: Pomacentridae) are a group of ecologically important, primarily coral reef fishes that include over 400 species. Damselfishes have been used as model organisms to study recruitment (anemonefishes), the effects of ocean acidification (spiny damselfish), population structure and speciation (Dascyllus). The genus Dascyllus includes a complex of relatively larger bodied species, the Dascyllus trimaculatus species complex. This complex is comprised of several species including D. trimaculatus itself. The three-spot damselfish, D. trimaculatus is a widespread, common, and well-studied coral reef fish species found across the tropical Indo-Pacific that has served as a model species for coral reef fish research. In this dissertation, we expand on what is known of population dynamics of the complex while providing valuable genomic resources to support future research into this model system. In Chapter 1 of this dissertation, we explore populations at the margins of a species’ distribution which tend to be smaller, more isolated, and importantly, exist in habitats at the edge of the species’ physiological limit. These conditions make such populations particularly vulnerable to extirpation, especially amid accelerating environmental changes. We focus on population structure based on mitochondrial DNA markers and inferred dynamics of an understudied, potentially vulnerable, population of an otherwise well studied and widespread coral reef fish. We present phylogeographic results using the most robust set of samples collected in the Ryukyu Archipelago and gain insight into edge population dynamics by comparing to another more well-studied edge population in French Polynesia. We found that despite its proximity to the Coral Triangle and position along the Kuroshio Current, the Ryukyu population of D. trimaculatus seems to be a relatively young and closed, suggesting that it may be more vulnerable to extinction than might be otherwise expected. In Chapter 2 we present the first genome assembly of Dascyllus trimaculatus. This assembly contains 910 Mb, 90% of the bases are in 24 chromosome-scale scaffolds, and the BUSCO score of the assembly is 97.9%. Our findings confirm previous reports of a karyotype of 2n = 47 in D. trimaculatus in which one parent contributes 24 chromosomes and the other 23. We find evidence that this karyotype is the result of a heterozygous Robertsonian fusion. This assembly was a necessary resource for research in Chapter 3 and will be a valuable resource in the population genomics and conservation of Damselfishes, and continued studies of the karyotypic diversity in this clade. In Chapter 3, we use low coverage whole genome sequencing to examine the population structure of its Pacific distribution, and again, focus on dynamics of the Ryukyu Archipelago in southern Japan. We find evidence of clear divergence between the Pacific and Indian Ocean populations as well as between the Pacific and the restricted edge population in the Society Islands at the eastern edge of its range. We see a clear signal of introgression of one species within the complex, D. auripinnis described from the Line and Phoenix islands into D. trimaculatus that extends across to the west Pacific - further than previously reported. We also find that there is directional but restricted gene flow of Philippine genotypes into the Ryukyu Archipelago. While we have not yet identified specific isolation and speciation mechanisms, we propose some likely scenarios. This dissertation builds upon previous research of this complex which had made it a model system in its own right and adds genomic resources to ensure that we continue to further our understanding of this important model system in this genomic age. There are now resources for endless directions and questions to follow regarding this species complex and evolutionary processes. Understanding mechanisms of dispersal, the role of selection, and the capabilities of local adaptation in a species with relatively restricted gene flow are some of the exciting questions that this project has uncovered, thus placing Dascyllus trimaculatus as an ideal model species to address fundamental questions in the ecology and evolution of coral reef fishes.

Overfishing is contributing to significant declines in marine species worldwide. Here, I evaluated the contribution of genetic data to enhancing conservation and fisheries management of commercially important fishes. First, I conducted two empirical population genetics studies on groupers that aggregate to spawn. I evaluated population structure of Nassau grouper (Epinephelus striatus) across the Caribbean Sea using mitochondrial DNA, microsatellites and single nucleotide polymorphisms (SNPs). Results suggested ocean currents and long distance migrations of adults to spawning sites may contribute to restricted gene flow and regional genetic differentiation among subpopulations, making it less likely that larvae settle beyond catchment areas of spawning sites. Regional patterns of genetic differentiation warrant international collaboration, standardizing fisheries management and conservation initiatives among countries within genetically isolated clusters. I then evaluated population structure of Leopard grouper (Mycteroperca rosacea) in the Gulf of California using mitochondrial DNA and microsatellites. I observed genetic differentiation and asymmetrical connectivity among subpopulations of M. rosacea, with divergence of lineages dating to the late Pleistocene and Holocene. Findings suggest a combination of marine reserves, catch quotas, and seasonal fishery closures as a potential management strategy for M. rosacea in the Gulf of California. Finally, I conducted a meta-analysis of 386 empirical genetics studies on marine and diadromous fishes in order to evaluate the implications of weak or strong population structure on the implementation of fisheries management. Quantitative adjectives used to describe varying magnitudes of genetic differentiation possessed statistically different mean F_ST values in studies utilizing mtDNA, microsatellites and allozymes. Additionally, mean F_ST values across stock management strategies were statistically different from one another. Results have major implications for standardizing reporting of metrics of genetic differentiation across empirical population genetics studies focusing on and prescribing stock management strategies for commercially important fishes.