<p> The macroevolutionary pattern of Rensch&rsquo;s Rule, i.e., positive allometry of sexual size dimorphism, has had mixed support in turtles. Using the largest carapace length dataset and the only large-scale body mass dataset heretofore assembled for this group, we determine (a) whether turtles conform to Rensch&rsquo;s Rule at the order, suborder, and family levels, and (b) whether inferences regarding allometry of sexual size dimorphism differ based on choice of body size metric used for analyses. We compiled large databases of mean body mass and carapace length for males and females of as many populations and species as possible using mostly primary literature. We then determined scaling relationships between males and females for average body mass and straight carapace length across species of turtles using traditional and phylogenetic comparative methods. We also used linear regression analyses to evaluate sex-specific differences in the variance explained by carapace length on body mass. </p><p> In non-phylogenetic analyses, body mass supports Rensch&rsquo;s Rule, whereas straight carapace length supports isometry. Using phylogenetic independent contrasts, both body mass and straight carapace length support Rensch&rsquo;s Rule with strong congruence between body size metrics. More variance is explained by mass than carapace length. At the family level, support for Rensch&rsquo;s Rule is more frequent when mass is used as a body size metric and in phylogenetic comparative analyses. Turtles do not differ in their mass-to-length regressions by sex. Turtles display Rensch&rsquo;s Rule overall and within some families of Cryptodires, but not in Pleurodire families. At broad scales, mass and length are strongly congruent with respect to Rensch&rsquo;s Rule in turtles, and discrepancies are observed mostly at the family level (which is the level where Rensch&rsquo;s Rule is most often evaluated). At macroevolutionary scales, the purported advantages of length measurements over weight measurements are not supported in these ectothermic vertebrates.</p>

<p>Billions of dollars have been invested in land protection as a strategy to conserve biodiversity based on the assumption that protected areas buffer species from processes that drive extinction. Increasingly, protected area expansion and connectivity are being incorporated into climate change adaptation strategies to facilitate anticipated shifts in species ranges in response to predicted changes in temperature and precipitation. However, the effectiveness of protected areas at maintaining biodiversity, either by reducing the risk of extinction or facilitating colonization into new areas, has not been well established. In addition, the growing reliance on multiple-use protected areas that allow resource extraction, such as timber harvest and mineral mining, has raised questions about whether multiple-use protected areas are equally beneficial for long-term biodiversity conservation as more strictly protected areas that limit active resource management. In order to address these questions using repeated Breeding Bird Atlas data, I first had to confront the limitations of existing approaches to account for imperfect detection by developing a novel modelling approach to addresses the gap between requirements of other multi-season occupancy models (i.e. repeated sampling) and existing datasets. I then applied that single-visit dynamic occupancy modelling approach to Atlas data in New York and Pennsylvania for 97 species to quantify drivers of colonization and extinction while accounting for imperfect detection in landscapes that varied by type and amount of land cover and area under protection. In general, protected areas increased colonization and lowered extinction probabilities to an increasing degree as both forest cover and neighborhood protection decreased, with particular benefits for forest breeding birds. Both strict and multiple-use protected areas increased colonization and reduced extinction more for mature forest species than early forest species, with the greatest benefits accruing when forest cover was relatively low. These results provided the most comprehensive evidence to date that protected areas can facilitate species persistence by both reducing the risk of extinction and providing attractive colonization sites as species? ranges shift and that biodiversity conservation can be compatible with renewable resource extraction.

<p> Rhamphichthyoidea is a monophyletic clade of Neotropical electric fishes (Ostariophysi, Gymnotiformes), with about 55 known species, some still awaiting formal description. These species are widespread in tropical and subtropical freshwaters of South America, with a single species reaching Costa Rica in southern Central America. Rhamphichthyoid species inhabit a variety of lowland habitats including small sandy streams, floodplain oxbow lakes, and the benthos of deep river channels. Rhamphichthyoidea displays substantial morphological disparity, ranging in maximum total body size from 62 millimeters to about 1 meter, and exhibiting diversity of distinctive snout shapes. This phenotypic diversity is reflected in a broad range of ecological interactions and foraging modes. Here I studied the diversity of Rhamphichthyoidea at three levels: alpha diversity, species interrelationships, and macroevolutionary patterns. The first chapter presents results of a species level phylogeny using a total evidence approach, including a new classification based on phylogenetically diagnosed clades. The second and third chapters review the species diversity and boundaries of the poorly studied genera <i>Rhamphichthys</i> (7 spp.) and <i>Gymnorhamphichthys</i> (5 spp.). The fourth chapter presents an analysis of diversification through time, linking habitat occupancy with morphological aspects of snout shape in the radiation of Rhamphichthyoidea. In summary, this dissertation provides information on the species diversity, morphological characteristics, biogeography of the long-snouted rhamphichthyids; the phylogenetic relationships and a newly proposed classification of Rhamphichthyoidea; and adds to the understanding of general patterns of diversification within the Neotropical ichthyofauna.</p>