<p> The integration of advances in computing technology with major innovations in sequence data collection and phylogenetic inference has revolutionized evolutionary biology in the 21^st century. In particular, the continual development of both theory and software that allow for more flexibility in utilizing molecular clock methods has radically transformed our understanding of the mode and tempo of diversification across the Tree of Life. Over the course of five chapters, this dissertation explores methodological challenges to phylogenetic inference with the aim of better understanding the evolutionary history of the Holocentridae (squirrelfishes and soldierfishes). </p><p> Chapter 1 begins by focusing on the problem of accommodating clade specific rate heterogeneity in molecular clock analyses. While various nucleotide substitution models have been developed to accommodate among lineage rate heterogeneity, recently developed "uncorrelated relaxed clock" and "random local clock" models are predicted to perform better in the presence of lineage specific rate heterogeneity as these models relax assumptions of inheritance of nucleotide substitution rates between descendant lineages. Using simulations and two cetacean (whale and dolphin) datasets as a case study, we demonstrate abrupt changes in rate isolated to one or a few lineages in the phylogeny can mislead rate and age estimation, even when the node of interest is calibrated; and provide suggestions for diagnosing extreme clade specific rate heterogeneity.</p><p> Homoplasy is another important, yet often overlooked, source of error in phylogenetic studies. Chapters 2 and 3 utilize phylogenetic informative approaches to screen nucleotide sequence data for homoplasious site patterns. Using phylogenetic informativeness profiles, chapter 2 reconciles two competing hypotheses of ray-finned fish divergence times by highlighting that mitogenomic based Jurassic and Triassic divergence time estimates for most major lineages of spiny-rayed (acanthomorph) fishes were an artifact of tree extension. Evolutionary relationships of early diverging acanthomorph fishes are also contentious, with molecular data supporting either holocentrids or a clade comprised of holocentrids and primarily deep-sea fishes as the sister lineage to the species-rich percomorpha. Chapter 3 reveals this conflict to also be largely driven by homoplasy and reconciles results based on previously published data with a 132 gene next-generation sequence dataset to identify the sister lineage of percomorph and the phylogenetic placement of holocentrid fishes.</p><p> Chapter 4 continues to explore holocentrid evolutionary relationships. Using a multi-locus dataset that includes all but one holocentrid genus, this chapter provides the first molecular phylogeny of the group. The systematics of holocentrid fishes has unstable for over 100 years. We demonstrate several of the key synapomorphies for holocentrid genera are in fact homoplasious. Likewise, several genera of holocentrine (squirrelfish) are rendered consistently paraphyletic by a series of maximum-likelihood and Bayesian analyses and we propose taxonomic revisions to reflect shared ancestry.</p><p> Chapter 5 further investigates the temporal history of holocentrid evolution. Contemporary holocentrid species richness is concentrated in the Indo-Australian Archipelago (IAA), yet these fishes also represent some of the most numerous fossil taxa in deposits of the Eocene West Tethyan biodiversity hotspot. Using likelihood-based methods integrated with a molecular timetree that incorporates fossils as tip taxa, we reconstruct the history of range evolution for these fishes. Following the collapse of the West Tethys, holocentrids exhibit a signature of increased range fragmentation, becoming isolated between the Atlantic and Indo-Pacific Ocean basins. However, rather than originating within the emerging IAA hotspot, the IAA appears to have acted as a reservoir for holocentrid diversity that originated in adjacent regions over deep evolutionary timescales. By integrating extinct lineages, these results provide a necessary historic perspective on the formation and maintenance of global marine biodiversity. </p>

<p> The Lesser Sunda Archipelago, also known as Nusa Tenggara, lies in the southeastern portion of Indonesia and extends between Bali in the west, and New Guinea in the east. While the Lesser Sundas themselves are oceanic islands that have never been land bridged to a continent the islands on either side do. Bali and the other Greater Sunda Islands of Java, Sumatra, and Borneo become periodically land bridged with Asia during glacial maxima forming the Sunda Shelf. New Guinea and Aru become periodically land bridged to Australia during glacial maxima and form the Sahul Shelf. Given their current orientation, the Lesser Sundas may act as &lsquo;stepping stones&rsquo; for animals and plants dispersing between the Sunda and Sahul Shelves and may act as a two-way filter for organisms dispersing between two of the world&rsquo;s great biogeographical realms. Alfred Russel Wallace&rsquo;s discovery of a pattern of clinal mixture of species from different biogeographical realms was a key insight leading to his identification of the Wallace Line and to his creation of the field of biogeography. Even though the Lesser Sundas played a critical role in the development of the field, this region has received little subsequent attention from historical biogeographers and our current understanding of Lesser Sunda biogeography has only modestly improved relative to what was known at the time of Wallace. The reptiles and amphibians of the Lesser Sundas represent a particularly interesting group of vertebrates from a biogeographical standpoint because they appear to show distributional patterns that are most consistent with a stepping-stone model of island colonization caused by the two-way filter zone. In Chapter 1, I review the geological and biogeographical literature for the Lesser Sundas and use these sources to formulate hypotheses concerning the colonization of the archipelago by rafting terrestrial vertebrates. In Chapters 2 through 4, I investigate the possibility that flying lizards, forest skinks, and fanged frogs have colonized the archipelago in a stepping-stone manner using a phylogenomic approach (using sequence data from mtDNA and hundreds of nuclear loci) whereby the relationships among island-specific lineages can be used to infer the sequence of island colonization. Flying lizards of the genus <i>Draco</i> form a monophyletic group that colonized the western Inner Arc islands of Lombok or Sumbawa from the Sunda Shelf around 10 million years ago when Lombok and Sumbawa first became land-positive. <i> Draco</i> continued expanding eastward through the Inner Arc until they reached Lembata, while a series of dispersal events from Flores south to Sumba, east to Timor, north to Wetar, west to Alor, and finally west to Pantar (the island immediately west of Lembata). The islands of Sumbawa and Flores contain multiple non-sister lineages that are parapatrically distributed and are exchanging migrants within an island. Forest Skinks of the genus <i> Sphenomorphus</i> show relatively little morphological divergence across their range yet exhibit large levels of genetic divergence. The oldest lineages of <i>Sphenomorphus</i> within the Lesser Sundas occur on the islands of Lombok and Flores and they expanded eastward through the Inner Arc until they reached Pantar. But rather than reaching Alor from neighboring Pantar, <i> Sphenomorphus</i> dispersed from Flores south to Sumba, then east to Timor, Alor, and Wetar. There are multiple non-sister lineages of <i> Sphenomorphus</i> on Lombok, Flores, and Sumba, and estimates of migration between lineages within each island suggest that these lineages are not interbreeding. Fanged frogs of the genus <i>Limnonectes</i> have colonized the Inner Arc of the Lesser Sundas from the Sunda Shelf. It is possible that <i> Limnonectes kadarsani</i> and <i>L. dammermani</i> diverged <i> in situ</i> on Lombok after which <i>L. kadarsani</i> dispersed east all the way to Lembata. But rather although a tree topology consistent with a stepping-stone pattern of island colonization is suggested by the mtDNA data, the phylogenomic results suggest a leap-frog pattern where Lembata is derived from West Flores, and these two lineages are closer related to Sumbawa than they are to Eastern Flores. The parapatrically distributed lineages on Flores are experiencing asymmetrical gene flow with successful migrants moving from west to east. In summary, the oldest islands of the western Inner Arc tend to harbor the most divergent lineages for all three focal taxa, a pattern expected from lineages originating from the Sunda Shelf. In <i>Draco </i> and <i>Sphenomorphus,</i> the islands of the eastern Inner Banda Arc are colonized by way of the &lsquo;Sumba Route&rsquo; where they disperse into the Outer Banda Arc island of Sumba and then move east to Timor, and finally north into the eastern Inner Arc. All three focal taxa show multiple non-sister lineages on some of the larger islands, suggesting either that multiple colonization events of a single island occurred, or possibly that formerly separated paleo-islands have since merged allowing for secondary contact of lineages that diverged in allopatry. These studies have shown that the biogeography of reptiles and amphibians within the Lesser Sundas is extremely complex. By examining biogeographical patterns across many co-distributed taxa these studies have the potential to provide insights into the geological history of the archipelago. From an evolutionary perspective, these studies highlight the presence of multiple independently evolving lineages within a currently described species occurring on the same island, which suggests that species diversity within reptiles and amphibians of the Lesser Sundas is underestimated.</p>