Search

This supplemental file contains methods for analyzing prior survey data along with a figure for that data. Additionally, it contains alpha and beta diversity figures for experiment one and relative abundance plots for individual tadpoles from both experiments.

Many endangered amphibian species survive in captive breeding facilities, but there have been few attempts to reintroduce captive-born individuals to rebuild wild populations. We conducted a soft-release trial of Limosa harlequin frogs, Atelopus limosus, which are highly susceptible to the amphibian chytrid fungus Batrachochytrium dendrobatidis (Bd), to understand changes associated with the transition from captivity to the wild. Specifically, we assessed changes in body condition, skin-associated bacterial communities and disease status after release. Frogs were housed individually in field mesocosms and monitored for 27 days. Body condition did not significantly change in the mesocosms, and was similar to, or higher than, that of wild conspecifics at day 27. The skin microbiome of captive-born frogs, based on 16S rRNA gene amplicons, became similar to that of wild frogs after 27 days in mesocosms. Prevalence of Bd in wild frogs was 13–27%, and 15% of the A. limosus in mesocosms became infected with Bd, but no mortality of infected frogs was observed. We conclude that mesocosms are suitable for systematically and repeatedly monitoring amphibians during release trials, and that body condition, the skin microbiome, and Bd status can all change within one month of placement of captive-born individuals back into the wild.

This file contains additional methods details and an experimental design figure.

Bacterial communities in and on wild hosts are increasingly appreciated for their importance in host health. Through both direct and indirect interactions, bacteria lining vertebrate gut mucosa provide hosts protection against infectious pathogens, sometimes even in distal body regions through immune regulation. In house finches (Haemorhous mexicanus), the bacterial pathogen Mycoplasma gallisepticum (MG) causes conjunctivitis, with ocular inflammation mediated by pro- and antiinflammatory cytokines and infection triggering MG-specific antibodies. Here, we tested the role of gut bacteria in host responses to MG by using oral antibiotics to perturb bacteria in the gut of captive house finches prior to experimental inoculation with MG. We found no clear support for an impact of gut bacterial disruption on conjunctival pathology, MG load, or plasma antibody levels. However, there was a non-significant trend for birds with intact gut communities to have greater conjunctival pathology, suggesting a possible impact of gut bacteria on pro-inflammatory cytokine stimulation. Using 16S bacterial rRNA amplicon sequencing, we found dramatic differences in cloacal bacterial community composition between captive, wild-caught house finches in our experiment and free-living finches from the same population, with lower bacterial richness and core communities composed of fewer genera in captive finches. We hypothesize that captivity may have affected the strength of results in this experiment, necessitating further study with this consideration. The abundance of anthropogenic impacts on wildlife and their bacterial communities, alongside the emergence and spread of infectious diseases, highlights the importance of studies addressing the role of commensal bacteria in health and disease, and the consequences of gut bacterial shifts on wild hosts. National Science Foundation [IOS-1755051, IOS-1755297]; NSF [IOS-1754872] Published version This work was supported by the National Science Foundation (IOS-1755051, IOS-1755297). Marissa M. Langager was supported by NSF grant IOS-1754872 to Dana M. Hawley. There was no additional external funding received for this study. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Vertebrate ocular microbiomes are poorly characterized and virtually unexplored in wildlife species. Pathogen defense is considered a key function of microbiomes, but determining microbiome stability during disease is critical for understanding the role of resident microbial communities in infectious disease dynamics. Here, we characterize the ocular bacterial microbiome of house finches (Haemorhous mexicanus), prior to and during experimental infection with an inflammatory ocular disease, Mycoplasmal conjunctivitis, caused by Mycoplasma gallisepticum. In ocular tissues of healthy house finches, we identified 526 total bacterial operational taxonomic units (OTUs, 97% similarity), primarily from Firmicutes (92.6%) and Proteobacteria (6.9%), via 16S rRNA gene amplicon sequencing. Resident ocular communities of healthy female finches were characterized by greater evenness and phylogenetic diversity compared to healthy male finches. Regardless of sex, ocular microbiome community structure significantly shifted 11 days after experimental inoculation with M. gallisepticum. A suite of OTUs, including taxa from the genera Methylobacterium, Acinetobacter, and Mycoplasma, appear to drive these changes, indicating that the whole finch ocular microbiome responds to infection. Further study is needed to quantify changes in absolute abundance of resident taxa and to elucidate potential functional roles of the resident ocular microbiome in mediating individual responses to this common songbird bacterial pathogen.

Host species vary in their propensity to become infected by and transmit parasites, and this variation in host competency can influence parasite transmission within host communities. Host competency is often attributed to morphological, physiological, and behavioral defenses of hosts, but hosts commonly have an additional, lesser studied form of protection: defensive symbionts. For instance, snails are facultatively defended by ectosymbiotic oligochaete worms (Chaetogaster limnaei) that consume free-living trematode parasites, bacteria protect amphibians from the fungus that causes chytridiomycosis, and ants protect plants from herbivores. In addition to reducing infection on their hosts, defensive symbionts may influence parasite transmission to other hosts by redirecting parasites toward other hosts and/or removing parasites from the system. We explored these possibilities by examining the relative roles of community composition and the presence of defensive symbionts (C. limnaei) in determining trematode infection intensity among second intermediate host communities composed of snails (Helisoma trivolvis) and tadpoles (Rana catesbeiana). Parasites were dramatically more successful at infecting snails than tadpoles, which led to more total parasites in host communities where snails were present. In addition, defensive symbionts substantially reduced snail infection intensity and thus reduced the total number of parasites in communities containing symbiont-defended snail hosts. Neither host community composition nor the presence of defensive symbionts on snails influenced individual tadpole infection in our experiments. Therefore, in our experiments, second intermediate host community structure did not influence individual host tadpole infection risk, but did influence total parasite transmission.

Due to advances in technology and data collection techniques, the number of measurements often exceeds the number of samples in ecological datasets. As such, standard models that attempt to assess the relationship between variables and a response are inapplicable and require a reduction in the number of dimensions to be estimable. Several filtering methods exist to accomplish this, including Indicator Species Analyses and Sure Information Screening, but these techniques often have questionable asymptotic properties or are not readily applicable to data with multinomial responses. As such, we propose and validate a new metric called the Kolmogorov-Smirnov Measure (KSM) to be used for filtering variables. In the paper, we develop the KSM, investigate its asymptotic properties, and compare it to group equalized Indicator Species Values through simulation studies and application to a well-known biological dataset.

Chytridiomycosis is a fungal skin disease in amphibians that is primarily caused by Batrachochytrium dendrobatidis (Bd). We analyzed whole genome sequences of 40 bacterial isolates that had been previously cultured from the skin of four amphibian species from Virginia, USA, and tested for their ability to inhibit Bd growth via an in vitro challenge assay. These 40 isolates spanned 11 families and 13 genera. The aim of this study was to identify genomic differences among the amphibian skin bacterial isolates and generate hypotheses about possible differences that could contribute to variation in their ability to inhibit the growth of Bd. We identified sixty-five gene families that were present in all 40 isolates. We also looked for the presence of biosynthetic gene clusters. While this set of isolates contained a wide variety of biosynthetic gene clusters, the two most abundant clusters with potential anti-fungal activity were siderophores (N=17) and Type III polyketide synthases (N=20). We then analyzed the isolates belonging to the phylum Proteobacteria in more detail. We identified 197 gene families that were present in all 22 Proteobacteria. We examined various subsets of the Proteobacteria for genes for specific compounds with known activity against fungi, including chitinase and violacein. We identified a difference in the number, as well as amino acid sequences, of predicted chitinases found in two isolates belonging to the genus Agrobacterium that varied in their inhibition of Bd. After examining the annotated genomes, we identified a predicted chitinase in a Sphingomonas isolate that inhibited the growth of Bd that was absent from the five Sphingomonas isolates that did not inhibit Bd growth. The genes vioA, vioB, vioC, vioD and vioE are necessary to produce violacein, a compound which inhibits the growth of Bd. Differences in these genes were identified in three out of the four Janthinobacterium isolates. Of these three isolates, two showed strong inhibition of Bd growth, while the third inhibited Bd growth to a lesser extent. Using comparative genomics, we generated several testable hypotheses about differences among bacterial isolates that could contribute to variation in ability to inhibit Bd growth. Further work is necessary to test the various mechanisms utilized by amphibian skin bacterial isolates to inhibit Bd. Master of Science Many amphibian population declines around the world have been caused by chytridiomycosis, a skin disease. This disease is caused by the fungus Batrachochytrium dendrobatidis (Bd). The skin of amphibians is also home to many bacteria that can provide important functions for the amphibian host, like preventing infection by Bd. To understand how these bacteria might provide protection, we examined the entire genomes of 40 bacterial isolates that reside on the skin of four amphibian species from Virginia, USA. These bacteria were previously tested for their ability to prevent Bd growth and 40 of them were chosen for sequencing based on selecting closely related isolates that varied in their ability to inhibit Bd growth. This allowed us to compare their genomes and generate hypotheses about possible genomic differences that could contribute to the variation in Bd growth inhibition. We identified sixty-five gene families that were present in all 40 bacteria. We also looked for sets of genes (biosynthetic gene clusters) that are known to produce secondary metabolites, which are compounds that can include antifungals. The two most abundant clusters we identified that had the potential to produce compounds that inhibit fungal growth were siderophores and Type III polyketide synthases. We then looked for genes that were not part of biosynthetic gene clusters that could produce specific compounds that can inhibit Bd growth, such as chitinase and violacein. We found variation in chitinase genes in several isolates that seemed to be associated with the ability to inhibit Bd growth. In addition, there were some differences in violacein genes that should be examined more in future studies. Overall, we suggest that using comparative genomic approaches can be valuable for identifying key bacterial functions in the microbiome.

Plant-associated microbes (collectively the microbiome) are important contributors to plant health. They are known to play roles in increasing yield via improving stress tolerance, promoting growth, and suppressing the activity of plant pathogens. We investigated the wheat seed-head microbiome (phytomicrobiome) as a potential source of resistance to Fusarium head blight (FHB), or scab. FHB is a devastating disease in wheat, and other cereal grains, that causes losses in both quantity, through reduced yield, and quality of grain, through the production of toxins such as Deoxynivalenol. Efforts to combat FHB have focused primarily on breeding cultivars with resistance and applying fungicides. However, new resources for combatting FHB may lie in microbiome-plant interactions. To explore host-microbiome-pathogen interactions, we used field trials to characterize the seed head bacterial community (16S rRNA gene amplicons) across planting locations, host resistance genotypes, varieties, and plant development stages. We identified bacterial amplicon sequence variants (ASVs) present in each sample and then examined ASV community composition based on our variables. Characterizing bacterial relative abundance across samples, we identified 9,063 ASVs. These ASVs clustered according to plant developmental stages or maturity plant, location, and host genotype, but not by variety or maturity group. First, comparing plants at the pre-flowering versus mature grain-head stage, we found that both bacterial community richness and evenness changed significantly. In addition to these developmental changes, we found that bacterial community structure changes across locations, even between locations. Finally, we found that, in the presence of the pathogen, ASVs cluster by host resistance genotype, and that there are important taxonomic groups that are differentially abundant in the presence of the pathogen. Overall, we found that the wheat grain-head microbiome is shaped by environment-host-pathogen interactions, and that these interactions lead to differential abundance of particular community members that may be important in the management of FHB. Master of Science in Life Sciences Plant associated microbes are important contributors to plant health. They are known to play roles in increasing yield via improved stress tolerance, promoting growth, and suppressing plant disease. We investigated the wheat grain-head microbial communities as a source of disease resistance. The disease is called Fusarium Head Blight (FHB) and is caused by Fusarium graminaerum. FHB is a devastating disease in wheat and other cereals, causing losses, through reduced yield and quality through the production of toxins that prohibit use of the grain. To combat FHB, research has focused on developing plants that have resistance and the application of chemical fungicides. However, new resources for combating FHB may lie in the interactions between plants and microbes. This research is focused on identifying microbes that naturally interact with the plant, and how the pathogen, Fusarium, interacts with these beneficial microbes. In field trials, we characterized the microbial community by DNA sequencing technologies across locations, wheat with varying levels of genetic resistance, and wheat developmental stages. First, between the wheat kernel samples of pre-flowering and maturity, we found significant differences in microbial community. Consistent with other studies we found that the largest changes in microbial community composition across different growing locations. Finally, we found an interaction between the grain head microbiome and host resistance state when plants were exposed to the pathogen. Overall, we find that the wheat grain head microbiome is shaped by growing location and through interactions with the plant host and pathogen.

Amphibians harbor beneficial skin bacteria that can contribute to host defense against chytridiomycosis, an infectious disease caused by the lethal fungal pathogen Batrachochytrium dendrobatidis (Bd). However, while skin-associated microbial communities may alter host infection risk, the structure of these complex microbial communities can be impacted by both biotic and abiotic factors. In a series of three studies, I investigated the natural temporal and spatial variation in bacterial communities on the skin of wild and captive-born amphibians using 16S rRNA gene amplicon sequencing to characterize bacterial community diversity. First, in a study examining the skin bacterial communities of two sympatric treefrog species (Agalychnis callidryas and Dendropsophus ebraccatus) at a single pond over multiple seasons and years, I found that annual, seasonal, and even daily fluctuations in temperature and rainfall changed the skin bacterial communities on these species. Second, I further investigated the impact of seasonality and rainfall on amphibian skin bacterial communities with a study of the bacterial communities on Craugastor fitzingeri, a common terrestrial species, along a rainfall gradient, and five co-occurring amphibian species at a single site. The strong wet and dry seasonality in the tropical lowland forest impacted the bacterial communities of multiple stream-dwelling co-occurring species, but the nature of the changes differed among the frog species. For C. fitzingeri sampled along the rainfall gradient, I found there was variation in bacterial community structure among sites, although this was not correlated with the latitudinal rainfall gradient. Finally, I investigated the challenges faced by captive-reared Atelopus limosus, an endangered amphibian species, after soft-release into natural habitat with the use of mesocosms. I found that the skin bacterial communities reverted to wild-type fairly quickly, body condition decreased to come closer to wild conspecifics, and 15% of the frogs became infected with Bd during the 27 day trial in mesocosms. Overall, I found that skin bacterial communities of lowland amphibians change across time and space, that variation sometimes correlates with environmental conditions at the time and the site of sampling, and that skin bacterial communities on captive-born frogs revert to wild-frog's state soon after soft-release to natural habitat. Doctor of Philosophy Beneficial bacteria found on amphibian skin can provide protection against an infectious disease caused by the lethal amphibian chytrid fungus (Batrachochytrium dendrobatidis), that has been linked with the decline and extinction of amphibian species worldwide. However, while skin bacterial communities may play a key role in determining disease outcome, these complex microbial communities can be impacted by biological and environmental factors. In a series of three studies, I investigated the natural variation in skin bacterial communities on wild and captive-born amphibians through time and space using modern DNA sequencing technologies to characterize bacterial community diversity. First, in a study examining the skin bacterial communities of two treefrog species at a single pond over multiple years and seasons, I found that annual, seasonal, and even daily fluctuations in temperature and rainfall changed the skin bacterial communities on these species. Second, I further investigated the impact of seasonality and rainfall with a study sampling the skin of one common frog species along a rainfall gradient, and five amphibian species at a single site across seasons. The strong wet and dry seasonality in the tropical lowland forest impacted the bacterial communities of multiple species found near streams, but the nature of the changes differed among the different frogs. For the common species sampled along the rainfall gradient, I found there was variation in bacterial community structure among sites, although this was not correlated with the rainfall gradient. Finally, I investigated the challenges faced by captive-reared Atelopus limosus, an endangered amphibian species from Panama, after release into field enclosures in the natural habitat. I found that the skin bacterial communities reverted to wild-type fairly quickly, body mass decreased to come closer to wild frogs of the same species, and 15% of the frogs became infected with the chytrid fungus during the 27 day trial in the field enclosures. Overall, I found that skin bacterial communities of lowland amphibians change across time and space, that variation is sometimes linked with environmental conditions at the time and site of sampling, and that captive-born frogs revert to wild states soon after release to natural habitat.

Anthropogenic resource provisioning of wildlife has increasingly been hypothesized to alter pathogen spread. Although bird feeding is the most widespread form of intentional wildlife provisioning, we know relatively little about how the degree of anthropogenic feeding at a site impacts wild birds in ways relevant to disease transmission. We manipulated the density of bird feeders (low versus high) available at otherwise similar sites and tracked the local abundance, body condition (scaled-mass index), feeding behavior, and movement across the landscape in wild house finches (Haemorhous mexicanus), a feeder-dependent species subject to outbreaks of a contagious pathogen commonly spread at feeders. The local abundance of house finches was significantly higher at sites with high feeder density but, surprisingly, finches at high-density feeder sites had poorer body condition than those at low-density sites. Behaviorally, birds at high-density feeder sites had longer average feeding bouts and spent more time per day on feeders than birds at low-density feeder sites. Further, birds first recorded at low-density feeder sites were more likely to move to a neighboring high-density feeder site than vice versa. Overall, because local abundance and time spent on feeders have been linked with the risk of disease outbreaks in this species, effects of bird feeder density on both traits may, in turn, influence disease dynamics in house finches. Our results suggest that heterogeneity in the density of bird feeders can have diverse effects on wild birds, with potential consequences for disease transmission. Master of Science Feeding wildlife has increasingly been thought to change the spread of disease. Although bird feeding is the most widespread form of intentional wildlife feeding, we know relatively little about how much human feeding impacts wild birds in ways that affect disease transmission. We changed the density of bird feeders (low versus high) available at otherwise similar areas and tracked the local abundance, body condition, feeding behavior, and movement across the landscape in wild house finches (Haemorhous mexicanus), a feeder-dependent species subject to outbreaks of a infectious disease commonly spread at feeders. The local abundance of house finches was significantly higher at sites with high feeder density but, surprisingly, finches at high-density feeder sites had poorer body condition than those at low-density sites. Behaviorally, birds at high-density feeder sites had longer average bouts on feeders and spent more time per day on feeders than birds at low-density feeder sites. Further, birds first recorded at low-density feeder sites were more likely to move to a neighboring high-density feeder site than vice versa. Overall, because local abundance and time spent on feeders have been linked with the risk of disease outbreaks in this species, effects of bird feeder density on both traits may, in turn, increase disease spread in house finches. Our results suggest that variation in the density of bird feeders can have diverse effects on wild birds, with potential consequences for disease transmission.

Emergent infectious diseases are a major driver of the accelerated rates of biodiversity loss that are being documented around the world. Global losses of amphibians provide evidence of this, especially those associated with chytridiomycosis, a lethal skin disease caused by the fungus Batrachochytrium dendrobatidis (Bd). Amphibian skin can harbor diverse bacterial communities that, in some cases, can inhibit the growth of Bd. Thus, there is interest in using skin bacteria as probiotics to mitigate Bd infections in amphibians. However, experiments testing this conservation approach have yielded mixed results, suggesting a lack of understanding about the ecology of these microbial communities. My dissertation research aimed to assess basic ecological questions in microbial ecology and to contribute to the development of probiotics using amphibian skin bacteria. First, to assess whether environmental conditions influence the function of amphibian skin bacterial communities, I conducted a field survey across low and high elevation populations of an amphibian host to assess their skin bacterial communities and metabolite profiles. I found that similar bacterial communities produced different metabolites at different locations, implying a potential functional plasticity. Second, since culturing is critical for characterizing bacteria, I aimed to identify the culture media (low vs high nutrient concentration) that recovers the most representative fraction of the amphibian skin bacterial community. I found that media with low nutrient concentrations cultured a higher diversity and recovered a more representative fraction of the diversity occurring on amphibian skin. I also determined that sampling more individuals is critical to maximize culture collections. Third, I assessed the diversity of the amphibian skin fungal community in relation to Bd infection across eight amphibian species. I determined that amphibian species was the most important predictor of fungal diversity and community structure, and that Bd infection did not have a strong impact. My dissertation highlights the importance of environmental conditions in the function of amphibian skin bacteria, expands our knowledge of the understudied fungal component of the amphibian skin microbiome, and complements current efforts in amphibian conservation. Ph. D. In light of the global losses of amphibian diversity due to, in part, the skin disease chytridiomycosis (caused by the fungus Batrachochytrium dendrobatidis [Bd]); the discovery that some amphibian-skin bacteria can inhibit Bd growth provides hope for amphibian conservation via their use as probiotics to control Bd infections. However, experiments testing these bacteria have yielded inconsistent results, suggesting a limited understanding about the factors influencing the diversity of amphibian-skin microbes and their ability to inhibit Bd. Also, efforts to identify effective candidates for probiotic therapy are still premature. Thus, my dissertation had an ecological emphasis and focused on complementing conservation efforts focused on probiotics. First, I assessed whether environmental conditions influence bacteriallyproduced products, which can have antifungal properties. Specifically, I surveyed low and highelevation populations of an amphibian species to assess the skin-bacteria and their products. I determined that, while skin bacterial communities were similar across an environmental gradient, their products differed, suggesting potential different antifungal properties. Second, I assessed the ability of different culture media types (low vs high nutrient concentrations) to grow a high portion and most representative fraction of the amphibian-skin bacteria. I found that culture media with low nutrient concentrations allowed the growth of a higher diversity of the bacteria occurring on the amphibian-skin, including the abundant members, and also determined that including a large number of amphibians is the best way to improve culture collections. Third, I assessed the fungal diversity occurring in the skin of different amphibian species and how it might response to Bd infections, and examined whether skin-fungi interact with co-occurring bacteria. I found that the amphibian species was the most important driver of the fungal diversity, and that Bd infection did not influence the diversity of these communities. Moreover, I identified the most diverse fungal phyla occurring in the amphibian-skin and determined that these fungi might interact with co-occurring bacteria. My dissertation contributes to our understanding about the influence of the environmental conditions in the amphibian-skin bacteria, expands our limited knowledge on the amphibian-skin fungi, and complement current amphibian conservation efforts.

Global change is altering the community composition, variability, and behavior of organisms in a diverse suite of ecosystems. Because of climate change and eutrophication, freshwater lakes and reservoirs are experiencing an increase in low dissolved oxygen concentrations (hypoxia) in their bottom waters (hypolimnion), which can disrupt ecological communities. Zooplankton, important aquatic organisms for regulating water quality and food webs, are one group of organisms affected by hypoxia since zooplankton need oxygen to respire. My research shows that hypoxia may disrupt zooplankton behavior and increase the variability of zooplankton communities. Zooplankton ubiquitously exhibit diel vertical migration, where the majority of the population resides in the hypolimnion during the daytime to escape predation from fish and damage from ultraviolet radiation. At night, many zooplankton ascend to the surface waters to feed on phytoplankton, when there is decreased risk of predation and radiation. My results from intensive 24-hour sampling campaigns suggest that hypolimnetic hypoxia may alter zooplankton migration, biomass, and behavior, which may in turn exacerbate water quality degradation due to the critical role zooplankton play in freshwater ecosystems. In addition, field surveys in four reservoirs over three years revealed that hypoxia may increase the variability of zooplankton communities compared to oxic conditions. Consequently, as lakes and reservoirs experience increased extent and duration of hypoxia in the future, it is critical to understand how more variable zooplankton communities alter freshwater ecosystem functioning. Ph. D.

Infectious disease is a ubiquitous aspect of life on earth; however, parasites and pathogens are not distributed equally among individual hosts. Due to its ability to shape the way that individuals interact with other potential hosts and the environment, behavior is one of the most salient ways through which host biology varies in the context of disease. Variation in animal behavior can impact both transmission and the extent of a host's pathogen acquisition, and thus can have important consequences for infectious disease dynamics. Additionally, in this world of rapid urbanization where landscapes and wildlife resources are being altered, it is important to understand the ways in which human activity impact wildlife behavior, and in turn, disease dynamics. Here, we used both observational and experimental studies in field and laboratory settings to investigate the relationships among host behavior and physiology, anthropogenic food sources, and disease transmission in a natural host-pathogen system. First, we examined the relationship between house finch (Haemorhous mexicanus) stress physiology, exploratory behaviors, and social behaviors in the wild. We provided evidence that more exploratory house finches interact with more individuals in the wild, and have higher baseline concentrations of circulating stress hormones. Next, we found evidence that the amount of time spent on bird feeders drives both the acquisition and transmission of the bacterial pathogen Mycoplasma gallisepticum (Mg), indicating that variation in host foraging behavior has important transmission consequences in this system. Lastly, we found that the density of bird feeders available to house finches predicts the extent of Mg transmission in captivity. Taken together, these results highlight the important role that behavioral heterogeneity can play in the acquisition and spread of pathogens, as well as the potential impacts of human behavior on wildlife disease dynamics. Future work should seek to identify specific physiological mechanisms driving Mg acquisition and transmission as they relate to variation in host behavior, and the ways in which bird feeders impact disease-relevant behaviors in the wild. Ph. D.

We live in an era of global change, where emerging infectious diseases such as Ebola, Zika, bird flu, and white nose syndrome are affecting humans, wildlife, and domesticated species at an increasing rate. To understand and predict the dynamic spread of these infectious agents and other symbionts through host populations and communities, we need dynamic mathematical models that accurately portray host-symbiont transmission. But transmission is an inherently difficult process to measure or study, because it is actually a series of interacting processes influenced by abiotic and biotic factors at multiple scales, and thus empirical tests of the transmission function within epidemiological models are rare. Therefore, in this dissertation, I explore factors at the individual, population, and community-levels that influence host contact rates or symbiont transmission success in a common snail-symbiont system, providing a detailed description of the multi-faceted nature of symbiont transmission. From a review of the ecological literature, I found that most models assume that transmission is a linear function of host population density, whereas most empirical studies describe transmission as a nonlinear function of density. I then quantified the net nonlinear transmission-density relationship in a system where ectosymbiotic oligochaetes are directly transmitted among snail hosts, and I explored the ecological mechanisms underlying the nonlinear transmission-density relationship observed in the field via intraspecific transmission success and contact rate experiments in the laboratory. I found that the field results could be explained by heterogeneity in transmission success among snails with different characteristics and nonlinear contact-density relationships caused by non-instantaneous handling times. After I 'unpacked'population-level transmission dynamics into those individual-level mechanistic processes, I used this same approach to examine higher-level ecological organization by describing the mechanistic underpinnings of interspecific or community-level transmission in the same snail-symbiont system. I found that low interspecific transmission rates in the field were the product of opposing interactions between high population densities, high prevalences of infection, and very low interspecific transmission success caused by strong symbiont preferences for their current host species. Unpacking transmission in this way resulted in one of the most detailed empirical studies of transmission dynamics in a wildlife system, and yielded many surprising new insights in symbiont ecology that would not have been discovered with a purely phenomenological or holistic view of transmission. Though simple, linear, and holistic epidemiological models will always be important tools in disease ecology, 'unpacking'transmission rates and adding heterogeneity and nonlinearity to models, as I have done here, will become increasingly important as we work to maximize model prediction accuracy in this era of increased disease emergence. Ph. D.

Understanding the ecological processes that impact parasite abundance and distribution is critically important for epidemiology and predicting how infectious disease dynamics may respond to future disturbance. Digenean trematodes (Platyhelminthes: Trematoda) are parasitic flatworms with complex, multi-host life cycles that include snail first-intermediate hosts and vertebrate definitive hosts. Trematodes cause numerous diseases of humans (e.g. schistosomiasis) and livestock (e.g. fascioliasis), and impact the ecology of wildlife systems. Identifying the ecological mechanisms that regulate these complex, multi-host interactions will advance both our understanding of parasitism and the dynamics of infectious disease. By examining patterns of infection in Elimia (= Oxytrema = Goniobasis) proxima snails, my dissertation research investigated the environmental factors and ecological processes that structure trematode communities in streams. First, I examined temporal variation in trematode infection of snails in five headwater streams. Over a three year period, I found no consistent seasonal patterns of trematode infection. There was consistency across sites in trematode prevalence, as sites with high prevalence at the beginning of the study tended to remain sites of high infection, relative to lower prevalence sites. Second, I examined landscape level variation in trematode infection by characterizing the regional distribution, abundance and diversity of E. proxima infections in 20 headwater streams. I found a broad scale spatial pattern in trematode communities due to regional turnover in dominant species. This pattern was correlated with elevation, but there were no significant relationships with other environmental variables. Additionally, molecular characterization of trematodes indicated the presence of cryptic (morphologically indistinguishable) species complexes within this system, and variation in genetic diversity among trematode types may reflect differences in host dispersal abilities. Third, I examined trematode infection within a single stream network across multiple headwaters and the mainstem. I found a decreasing downstream gradient of trematode prevalence related to several environmental variables including elevation, snail density, conductivity, and stream depth. Additionally, headwater communities were nested subsets of the communities found in the mainstem. By combining approaches at different temporal and spatial scales, my dissertation research increases our understanding of the processes that impact the abundance and distribution of parasites. Ph. D.

Phytoplankton populations are integral to the structure and function of aquatic ecosystems, and phytoplankton are an excellent study system for exploring ecological questions. Reservoirs often exhibit high horizontal (inflow to dam) and vertical (surface to sediments) environmental heterogeneity, which plays a large role in determining phytoplankton population dynamics. In this thesis, I explore how three bloom-forming phytoplankton taxa, the dinoflagellates Peridinium and Gymnodinium, and the cyanobacterium Planktothrix, respond to horizontal and vertical environmental gradients, respectively. First, I monitored recruitment, or the process of leaving the sediments and entering the pelagic life stage, of dinoflagellates across a horizontal reservoir ecosystem gradient. Surprisingly, coupling of dinoflagellate biology with reservoir physics and chemistry varied along this continuum; recruiting cells were sensitive to reservoir physics (e.g., flow rate, solar radiation) in the upstream riverine zone, while recruitment was related to reservoir chemistry (e.g., dissolved oxygen, nutrients) in the downstream lacustrine zone. This study indicates that upstream habitats should be monitored when studying reservoir phytoplankton dynamics. Next, I investigated the environmental drivers of the vertical distribution and biomass of a hypolimnetic cyanobacterial bloom over two consecutive summers. I collected high-resolution in situ phytoplankton data, and measured environmental variables throughout the water column. Across both years, the vertical distribution of this population was determined by light availability, while the cyanobacterial biomass was predicted by both light and nutrients. These two studies demonstrate that changing physics and chemistry across environmental gradients can regulate phytoplankton dynamics in reservoirs, and phytoplankton monitoring should include more spatially comprehensive sampling approaches. Master of Science

Understanding how species respond to anthropogenic changes and why species respond in the way that they do can help focus conservation planning. Hellbenders (Cryptobranchus alleganiensis) are a freshwater species of increasing conservation concern that are suspected to have declined due to loss of forest cover. However, quantitative evidence of land use effects on hellbenders is lacking. I used a multilevel approach to understanding whether hellbenders respond to land use by examining physiological indices, demographics and patterns of species occurrence as endpoints. My study area included two major river drainages in Virginia which mark a largely understudied portion of the species' range. In Chapter I, I described hellbender distribution and tested the hypothesis that hellbenders would be more likely to occur in heavily forested landscapes. Surprisingly, hellbenders occupied a relatively wide land use gradient (range = 50-90% forest in an upstream catchment) and current land use was an unreliable predictor of occurrence. In Chapter II I examined hellbender abundance and demographics at a subset of study sites stratified across a land use gradient. Abundance of sub-adult/adult hellbenders increased as forest cover increased in collective upstream riparian areas, primarily as the result of increased recruitment of new adults to local populations (rather than increased apparent survival of adults). Populations in lesser forested areas were declining and composed largely of relatively old adults, indicating that land use can lead to changes in hellbender distribution given sufficient time. In Chapter III, I examined three indices of physiological condition (body condition, hematocrit and leukocyte profiles) in hellbenders captured across a land use gradient. I found evidence suggesting low reproductive success may explain reduced recruitment in areas of low forest cover and evidence suggesting hellbender endocrinology during the breeding season may vary with land use. In Chapter IV I examine efficacy of artificial shelters as a monitoring tool and demonstrate their potential as tool to further our understanding of mechanisms underlying demographic responses of hellbenders to land use. I synthesize my findings in Chapter V and conclude that loss of forest cover in riparian areas poses a formidable threat to hellbender population viability in Virginia. Ph. D.

Identifying the processes and mechanisms that govern communities of organisms is the main goal of community ecology. Locally operating mechanisms such as environmental filtering, in which the environment determines what species are found in a given location, as well as regional processes such as dispersal have all been identified as potential drivers of community processes. However, the relative importance of these drivers may vary temporally and spatially. In dendritic stream networks, headwater streams are isolated when compared to more centrally located mainstem stream sections. I investigated the potential for stream networks to influence the relative influence of local and regional processes via a survey and field experiment based approaches. I found that headwater streams can influence mainstem stream communities, potentially as a result of the dispersal of organisms or abiotic materials. Additionally, I demonstrated that macroinvertebrate communities in headwater streams respond more strongly to manipulations of local environment than do mainstem streams, both in terms of community composition as determined taxonomically and as functional traits. These results indicate that headwater streams may be affected differently than mainstem streams by anthropogenic activity and as such, management strategies and restorations may need to be specifically tailored to address the relative influences of local and regional processes at varying points within a stream network. Ph. D.

As data collection techniques improve, oftentimes the number of covariates exceeds the number of observations. When this happens, regression models become oversaturated and, thus, inestimable. Many classical and Bayesian techniques have been designed to combat this difficulty, with various means of combating the oversaturation. However, these techniques can be tricky to implement well, difficult to interpret, and unstable. What is proposed is a technique that takes advantage of the natural clustering of variables that can often be found in biological and ecological datasets known as the omics datasests. Generally speaking, omics datasets attempt to classify host species structure or function by characterizing a group of biological molecules, such as genes (Genomics), the proteins (Proteomics), and metabolites (Metabolomics). By clustering the covariates and regressing on a single value for each cluster, the model becomes both estimable and stable. In addition, the technique can account for the variability within each cluster, allow for the inclusion of expert judgment, and provide a probability of inclusion for each cluster. Ph. D.

Life-history strategies are adaptations in behavior, physiology, and anatomy that influence survival and reproductive success. Variation in life-history strategies is often determined by adaptations to environmental conditions and trade-offs with sexually-selected signals. One of the aspects controlling life-history trade-offs is the endocrine system. Testosterone is a hormone that mediates several key aspects of male reproduction, yet little is known about the causes and consequences of variation in testosterone. Using rufous-collared sparrows (Zonotrichia capensis), a Neotropical songbird with a wide distribution, I explored geographical patterns of variation in testosterone levels and infection by haemosporidians, a type of blood parasite. I found that testosterone did not vary with elevation, nor predict haemosporidian infection, but males in breeding condition were more likely to be infected (Chapter I). High levels of testosterone have been associated with an increased number of sexual contacts and can suppress the immune response, thus it may increase the risk of sexually transmitted infections. By studying the communities of bacteria that reside in the cloaca of birds, I found that they were different depending on testosterone levels, and that high-testosterone males had higher relative abundance of Chlamydiae, a class of intracellular pathogens (Chapter II). During the breeding season there is an increase in physical contacts among individuals, testosterone levels increase in males, and there are additional energetic demands, all of which can increase exposure to bacteria or facilitate infection. I compared the cloacal microbiome of the same individuals between breeding and non-breeding seasons, and found that in males, but not in females, bacterial richness and phylogenetic diversity increased when birds were in reproductive condition. This suggested that the cloacal microbiome in birds is dynamic and responsive to breeding condition and sex of the host (Chapter III). Lastly, I synthesized the most relevant findings and suggested directions for future work (Chapter IV). I conclude that variation in testosterone is not always associated with immune suppression, and that the links among reproductive physiology, behavior, and the microbiome can provide insight into the evolution of life-history strategies. Ph. D.

Invasive plants can displace native species, deteriorate native forest, and change plant communities and ecosystem functions. Native plant populations are fundamentally impacted by invasive species because of the interactions between invasive species and native plants. This study focuses on understanding the extent, mechanisms and consequences of interaction between a non-indigenous invader Ailanthus altissima and its functionally similar native species Robinia pseudoacacia in the Mid-Appalachian region, from an individual scale to a regional scale. These two subject species are common and coexist in early-successional eastern deciduous forest. The interactions between these two common species are important to community structure and canopy tree regeneration. To address the type and extent of interactions of these two species, a greenhouse experiment utilizing various species proportions, nutrient levels and seed sources was performed. In addition, a common-garden experiment with various species densities and proportions over three consecutive growing seasons was performed in a more natural condition than that of the greenhouse experiment. We found at the seedling stage, the dominant interaction was competition, and R. pseudoacacia was the winner both above- and belowground. The allelopathic compounds of A. altissima may have inhibited nodulation of R. pseudoacacia. Ailanthus altissima seedlings from its native region had slightly stronger competitive abilities compared with the seedlings from its invaded range. In the common garden experiment, R. pseudoacacia plants grew quicker than A. altissima, but A. altissima inhibited the growth of R. pseudoacacia by interspecific competition. The negative impact of A. altissima on R. pseudoacacia became larger as time progressed. To assess the community-level consequences of the two species, we conducted a forest mapping and a complete target-tree-based forest survey, and analyzed regional-scale data from the Forest Inventory Analysis Data Base. The two target species were significantly associated with themselves and with each other. Community species composition and diversity were significantly different across sites. A negative impact of both species on the understory community diversity and tree regeneration at the neighborhood scale was detected; while at a regional level, tree diversity in the FIA plots with either A. altissima or R. pseudoacacia was higher than the reference plots. Ph. D.

Crayfish throughout North America, Europe, and Asia host assemblages of obligate ectosymbiotic annelid worms called branchiobdellidans. The work presented here is a detailed experimental and observational study of the ecological interactions between crayfish and their worms. In a comprehensive literature review, I show that branchiobdellidans have complex and context-dependent effects on their hosts, serving as both beneficial cleaners and tissue-consuming parasites. Using a field survey and laboratory experiments, I provide novel evidence for age-specific resistance as an adaptation to maximize life-long benefits of a mutualism. Specifically, I show that Cambarus crayfish display a consistent ontogenetic shift in resistance to the colonization of branchiobdellidans and this shift likely reflects underlying changes in the costs and benefits of symbiosis. I then show that this change in host resistance creates predictable patterns of symbiont diversity and composition throughout host ontogeny. Host resistance limits within-host symbiont communities to a few weakly interacting species, whereas relaxed resistance leads to more diverse symbiont communities that have strong interactions among symbiont taxa. Thus, host resistance has direct effects on within-host symbiont community structure by selectively filtering colonizing species, and indirect effects by moderating the strength of interactions among symbionts. Lastly, in a detailed study of the worm Cambarincola ingens, I depict a symbiont dispersal strategy that yields highly predictable transmission dynamics during pairwise host-host encounters and shows that variation in transmission dynamics can be explained by the fitness outcomes for dispersing symbionts. Field observations revealed that worm reproduction is contingent on host size and intraspecific competition for preferred microhabitats. Using a predictive model that assumes transmission of symbionts only when current conditions yield fitness below a minimum threshold, I was able to predict individual transmission events much more accurately than a comparable null model that assumed a fixed probability of transmission. My work provides empirical support for the emerging trend among researchers that advocates the adaptation of general ecological frameworks to understand symbiont population structure and diversity, but my work also emphasizes the value of detailed natural history studies to uncover system-specific ecological and co-evolutionary processes such as partner control mechanisms, symbiont microhabitat selections, and symbiont dispersal strategies. Ph. D.

Amphibian populations have undergone unprecedented declines in recent decades. Many of these declines are due to the spread of the cutaneous fungal pathogen Batrachochytrium dendrobatidis (Bd), which causes the disease chytridiomycosis. The Panamanian golden frog (Atelopus zeteki) has not been seen in the wild since Bd spread through western Panama. In response to initial declines, golden frogs were collected from wild populations and placed in captive colonies with the goal of future reintroductions. An understanding of this species' natural defense mechanisms against Bd is needed for reintroduction to be successful. Previous studies indicate that cutaneous bacteria are an important defense mechanism for some amphibians and applying antifungal bacteria to the skin of Bd-susceptible amphibians (probiotic therapy) can prevent chytridiomycosis. Therefore, the goals of my dissertation were to characterize the bacterial community of A. zeteki and determine if probiotic therapy could be used to prevent chytridiomycosis in this species. I initially characterized the bacterial community of wild and captive golden frogs using samples collected prior to the initial declines and after approximately eight years in captivity. I found that the community structure of the microbiota was significantly different between wild and captive frogs; however, the offspring of the original captive frogs still shared 70% of their microbial community with wild frogs. Then, I characterized the Bd-inhibitory properties of 484 bacteria isolated from 11 species of free-living Panamanian amphibians. I found a large proportion of bacteria (75.2%) had the ability to inhibit Bd and this trait was widely distributed among bacterial taxa, although there was also significant variation within bacterial genera in their ability to inhibit Bd growth. I then experimentally tested the ability of four of these isolates to prevent chytridiomycosis in captive golden frogs. None of them successfully prevented infection; however, there were several frogs that cleared infection and this was correlated with composition of the bacteria initially present on their skin. Overall these results demonstrate that the structure of microbial communities of A. zeteki are important to host health and building on this might provide the best hope for reintroducing this iconic species back to its native habitat. Ph. D.

As part of the ongoing loss of global biodiversity, amphibian populations are experiencing declines and extinctions. A primary factor in these declines is the skin disease chytridiomycosis, which is caused by the fungus Batrachochytrium dendrobatidis (Bd). Recent research suggests that the amphibian skin microbiota has anti-Bd activity and may be an important factor in host disease resistance. However, little is known about the basic ecology of this host-microbe symbiosis, such as how much variation there is in microbial symbionts among host species and populations, and the nature of symbiont transmission, culturability, and function. My dissertation research addressed these basic questions in microbial ecology, as well as used a novel system to examine the long-standing ecological theory of community structure-function relationships. First, host-specificity, population-level variation and potential environmental transmission of the microbiota were examined by conducting a field survey of bacterial communities from bullfrogs, newts, pond water, and pond substrate at a single pond, and newts from multiple ponds. There was variation among amphibian host species and populations in their skin symbionts, and, in a host species-specific manner, amphibian skin may select for microbes that are generally in low abundance in the environment. Second, the culturability of amphibian skin bacteria was assessed by directly comparing culture-dependent and -independent bacterial sequences from the same individuals. Although less than 7% of the amphibian skin microbes were captured using R2A medium, most of the dominant bacteria were represented in our cultures, and similar patterns of diversity among four amphibian species were captured with both approaches. Third, the relationship between microbial community structure and function and selective forces shaping structure and function were examined in bullfrogs by tracking microbial community structure and function following experimental manipulation of the skin microbiota and pathogen exposure. Results of this study demonstrated that Bd is a selective force on cutaneous bacterial community structure and function, and suggest that beneficial states of bacterial structure and function may serve to limit infection and negative fitness consequences of Bd exposure. Using a combination of observational and experimental approaches, my dissertation contributes to understanding structure-function relationships of these complex symbiotic communities of vertebrates. Ph. D.

Bathyplectes anurus (Thomson) and B. curculionis (Thomson) (Hymenoptera: Ichneumonidae) are specialist parasitoids introduced to the United States for classical biological control of alfalfa weevil, Hypera postica (Coleoptera: Curculionidae). Adults of both species are morphologically similar, generally described as wasps ≈3-4 mm long with black, robust bodies. The Bathyplectes spp. occupy sympatric ranges throughout much of the United States and often coexist within alfalfa fields. In Virginia, B. anurus is the primary biological control agent, accounting for more than 90% of alfalfa weevil parasitization. A previous study, however, found that the levels of control by B. anurus varied among the three alfalfa-growing regions of Virginia, and noted that little is known about the mechanisms or factors responsible for the differential levels of parasitization. Of the factors that may affect host parasitization, three were addressed in this study: population density and spatial distribution of host and parasitoid, the effectiveness of parasitoids in response to host density as measured by the population functional response, and genetic variation among and between populations of parasitoids. GPS-referenced quadrat sampling was conducted in alfalfa fields in the Piedmont and southwestern region to assess alfalfa weevil population levels and spatial distribution, and the level and distribution of parasitization by B. anurus and B. curculionis. Results showed that there was no significant difference in alfalfa weevil densities among alfalfa fields in the Piedmont and southwestern region, and also no significant difference between regions in the level of parasitization. Mean alfalfa weevil density per 0.2-m2 quadrat per sampling date was 11.0 in the Piedmont and 8.0 in the southwestern region; mean percent parasitization per 0.2-m2 quadrat per sampling date was 9.1% in the Piedmont and 9.9% in the southwestern region. The spatial distributions of alfalfa weevil larvae and parasitization were found to be aggregated in all alfalfa fields sampled. The population functional response of B. anurus attack on alfalfa weevil was determined qualitatively to be Type II in all alfalfa fields sampled, except the field located at Blacksburg in the southwestern region. The Type II functional response suggests that, in general, the parasitoid has an inverse density dependent attack response to the density of alfalfa weevil larvae in the two regions of Virginia. Molecular analyses were conducted to 1) derive genetic sequences for B. anurus and B. curculionis, which could be used to identify and differentiate the two species, and 2) test for genetic variation within and between regional populations of Bathyplectes. Sequences for each species were obtained for the D2–D3 region of the 28S ribosomal DNA (28S rDNA) gene; sequence identity between B. anurus and B. curculionis was 96.6%. The 28S rDNA sequences were also obtained for another Bathyplectes species, possibly B. stenostigma, and for four taxa of Hymenoptera that hyperparasitize B. anurus and B. curculionis cocoons. Sequences were obtained for B. anurus and B. curculionis for a portion of the mitochondrial gene cytochrome oxidase subunit 1 (COI); sequence identity between B. anurus and B. curculionis was 89.7%. Genetic variation in the COI gene was observed for both species. One variation of the COI sequence expressed by 25% of B. anurus specimens (Haplotype 2) might indicate regional genetic variation correlated to temperature, in particular the average number of days per year that a region experiences temperatures above 32°C. Two major COI sequence haplotypes found in B. curculionis specimens might be indicative of incipient speciation. Ph. D.

Metagonimoides oregonensis is a digenetic trematode that infects raccoons as definitive hosts, the snail Elimia proxima as a first intermediate host and in the southern Appalachians, encysts in the muscle tissue of a variety of second intermediate salamander hosts. My first study examined 289 individual salamanders representing six species from 23 streams in North Carolina to determine which species of salamanders are naturally infected. I found that five of the six species examined had natural infections, but that there was variation in infection intensity and prevalence among the species. Of the six species, Desmognathus quadramaculatus may be most important in transmission, as they had the highest prevalence and intensity of infection. This may be due to their long larval period, which results in a longer trematode accrual period. My second study explored the role of host and parasite behavior in driving infection dynamics in this system. I examined both parasite response to host chemical cues and host response to parasite presence and chemical cues. I did not see a behavioral response by either the parasite or the host, indicating behavior is probably less important in determining variable infection among hosts in this system, than are environmental and ecological factors. My final study examined the effect of cercariae exposure (n=0, 20, 60) on locomotor performance of D. quadramaculatus, Eurycea wilderae and Hyla versicolor. I did not see any effect on locomotor performance for any of the species. Master of Science

Competing paradigms of community assembly emphasize different mechanisms for predicting patterns in biogeography. Niche assembly emphasizes the role of environmental gradients as filters that organize a metacommunity by locally selecting colonizers with similar functional traits, whereas dispersal assembly emphasizes the importance of source pool characteristics and dispersal limitation in organizing a metacommunity. In this study, I developed a framework that uses spatially explicit patterns in taxonomic and functional measures of community composition as diagnostics for community assembly processes for benthic macroinvertebrates in headwater streams in the southern Appalachians. Distance decay in taxonomic and functional similarity was used to determine the scales at which taxonomic turnover occurred within functional niches. Trait-neutral models of community composition were used as null models to assess which functional traits were the best candidates to explain how community composition was influenced by environmental gradients: an assessment of niche-based community assembly. Regional scale patterns suggested that niche-based community assembly was the dominant mechanism organizing community composition in headwater streams at local scales (

The family Bunyaviridae is the largest of vertebrate diseases and includes the mosquito-borne disease La Crosse (LAC) virus. Vectors include the major vector Aedes triseriatus and two accessory vectors Aedes albopictus and Aedes japonicus. In the past several decades there has been an increase in the number of LAC cases, implication of new vectors, and the appearance of new foci of disease in the southeastern U.S. To better understand the vectors and the relationship between vectors and the virus, laboratory and field studies were conducted to determine oviposition preferences, effect of virus infection on blood-feeding behavior, and compare the efficacy of various methods to determine minimum infection rates of vectors. In laboratory studies of oviposition preference, only Ae. japonicus demonstrated a preference when presented with preexisting eggs. They deposited more eggs in cups containing either conspecifics or Ae. albopictus. The presence of 1st instar larvae Ae. albopictus larvae deterred oviposition by Ae. triseriatus and Ae. japonicus. Ae. japonicus and Ae. triseriatus preferred cups containing larval rearing water (LRW) of conspecifics and Ae. albopictus. Aedes albopictus preferred LRW regardless of species compared to control cups. Field experiments with fresh egg papers and preexisting eggs did not show significant differences, although the unequal population densities of species in the study area confounded the analysis. More work is needed to elucidate the interaction among these species and its effect on oviposition in the field. Blood-feeding experiments showed that LAC virus-infected Ae. triseriatus and Ae. albopictus imbibed significantly less blood compared to uninfected mosquitoes. Because blood meal size affects the subsequent inhibition of host seeking, experiments were done to ascertain the effect of virus infection on refeeding. Significantly more infected Ae. triseriatus mosquitoes refed but there was no effect on the refeeding rate of Ae. albopictus. Thus, the detrimental effect of virus infection, i.e., reduction in blood meal size, may lead to increased host exposure by Ae. triseriatus, enhancing horizontal transmission. Collecting adult mosquitoes was more efficient to detect virus in field populations than the collection of eggs. Maximum likelihood estimation-infection rates (MLE-IR) were calculated using bias-corrected maximum likelihood estimation. Adult collections yielded significantly more positive pools compared with egg collections. Virus was isolated from pools from Ae. canadensis, Ae. triseriatus, and Ae. albopictus. These results are comparable to other studies. Ph. D.

Mountaintop removal and valley fill (MTRVF) mining has become a widespread means of coal extraction in the central Appalachians. During MTRVF several hundred meters of overburden are removed to access coal seams, and excess rubble is dumped into adjoining valleys and streams. Filling valleys eliminates stream headwaters and may result in loss of stream ecosystem functions, which are dependent on temporal and lateral connectivity in river networks. To determine the affect of MTRVF on stream ecosystem function, leaf breakdown, which is an ecosystem level attribute of forested streams, was measured in five streams draining MTRVF sites and five reference streams in central West Virginia. Leaf packs of white oak and red maple were installed in these streams in December 2007, leaves were collected in January, February, March, April, and June of 2008, and leaves were washed and processed in the lab. Leaf breakdown rates were significantly slower in filled streams. MTRVF streams were marked by high sediment levels, elevated base flow, elevated conductivity and pH, and a lower density and richness of shredding macroinvertebrates than reference sites, suggesting that slower leaf decay was the result of the combined set of altered conditions in MTRVF streams. Additionally, MTRVF streams showed no species-level difference between red maple and white oak breakdown rates, indicating that MTRVF inhibits control of ecosystem function exerted by leaf species characteristics. Master of Science

Anthropogenic disturbances can alter the abiotic composition of freshwater systems. These compositional changes can act as physiological stressors towards system inhabitants. However, little is known about how these altered abiotic factors influence stress hormones (corticosterone) in larval amphibians. Throughout the following chapters, I examined the effects of several abiotic factors on baseline and stress-induced corticosterone levels in the larvae of four amphibian species: Jefferson salamander (Ambystoma jeffersonianum), spotted salamander (A. maculatum), wood frog (Rana sylvatica), and grey treefrog (Hyla versicolor). Chapter II examined corticosterone level differences throughout development in A. jeffersonianum and R. sylvatica larvae under field, mesocosm, and laboratory venues. Baseline corticosterone levels in R. sylvatica increased near metamorphic climax in all venues, but not in A. jeffersonianum. Rather, baseline corticosterone levels differed with respect to venue throughout development in A. jeffersonianum. Chapter III examined corticosterone level differences among free-living A. jeffersonianum populations and possible abiotic factors underlying these hormone differences. Corticosterone levels significantly differed across populations. Increased baseline corticosterone levels significantly correlated to low pH. There was also a trend for increased baseline corticosterone levels to be positively correlated with chloride levels and negatively correlated with conductivity. Chapter IV examined the effects of laboratory manipulated pH on corticosterone levels in A. jeffersonianum, A. maculatum, R. sylvatica, and H. versicolor. There was a significant correlation between increased baseline corticosterone levels to low pH in all four species. Prey consumption (in both Ambystoma species) and survival (in A. jeffersonianum, A. maculatum, and R. sylvatica) were also negatively correlated to low pH. Chapter V examined the effects of increased conductivity on corticosterone levels in A. jeffersonianum, R. sylvatica, and H. versicolor. Increased conductivity exposure significantly correlated to increased baseline corticosterone levels in A. jeffersonianum and R. sylvatica. Prey consumption in A. jeffersonianum was also negatively correlated to increased conductivity. My dissertation shows that abiotic factors, such as pH and conductivity, can influence corticosterone levels in larval amphibians. These results suggest that corticosterone levels in larval amphibians may be a suitable biomarker reflective of altered freshwater habitat quality. However, my results also suggest that one should use a high degree of caution when using corticosterone levels in larval amphibians as a means to infer the health status of a population. Ph. D.

Island biogeography theory predicts that the patterns and distributions of spatially isolated populations are governed by large scale processes. The high elevations forests in the Southern Appalachians represent a series of naturally fragmented islands that harbor many isolated populations of species at the southern limits of their range. Understanding the governing forces of population dynamics in this region will enhance the probability of species persistence in the face of threats such as global warming and human development. We surveyed bird populations across multiple elevations in Virginia and combined this with a multi-scale habitat analysis to determine influences of species presence and species richness. We detected 101 species across the elevation gradient, including 12 species with special conservation status and ten species whose presence increased with increasing elevation. These ten elevation sensitive species responded to habitat variables at both the microhabitat and landscape scale, with species-specific patterns of habitat variable correlation emerging. Habitat type was least effective in predicting species presence for any elevation sensitive species. Species richness declined over the elevation gradient until the highest elevations, where this trend reversed and richness began to increase. This pattern was driven by an increase in short-distance migrants beginning at mid-elevations, which ultimately overpowered a corresponding decrease in long-distance migrants beginning at similar elevations. Habitat analysis linked these patterns to a preference of short-distance migrants for smaller, more isolated non-forested patches, and a historical lack of persistence for long-distance migrants. Conservation and management decisions for the region should focus on a multi-scale approach that preserves all habitat types for continued species presence and high species richness, although the persistence of particular elevation sensitive species is compounded by unique species-habitat relationships and the perception of islands as species-specific. Continued monitoring of these fragmented populations in light of both short- and long-term threats which span multiple scales of influence will maintain high species richness and ensure the persistence of crucial breeding habitat. Master of Science

To examine the interactions of disease and pollution on amphibian populations around the world, I investigated the effect of infection on contaminant susceptibility in pickerel frog, Rana palustris, larvae. I conducted standard 48-hr toxicity tests to examine effect of trematode parasite, Echinostoma trivolvis, infection (0, 10, or 30 cercaria) on the susceptibility of pickerel frog tadpoles to the widely used organophosphate insecticide malathion. LC50 values ranged from 16.5 – 17.4 mg/L, within the range reported for other amphibian species. I found no differences in susceptibility to malathion among parasite treatments. Nevertheless, this crucial question remains to be tested in other amphibian host-parasite systems. Second, I studied the reverse interaction, the effect of pesticide exposure on susceptibility to parasite infection. I exposed pickerel frog embryos to low doses of malathion, then subjected morphologically normal tadpoles to E. trivolvis later in development. Malathion significantly decreased hatching success and viability rates at concentrations lower than previously documented for anuran embryos. After 7 wk of development in water with no malathion, tadpoles previously exposed to malathion as embryos suffered increased parasite encystment rates compared to controls. My research identifies embryonic development as a sensitive window and the potential for increased susceptibility to infection long after pesticide exposure has ceased. With potential for increased parasite prevalence from eutrophication and climate change, my data underscore the importance of understanding the reciprocal influences of parasites and pesticides in amphibians. Master of Science

The widespread use and subsequent release of pesticides into aquatic environments have sparked concerns about how organisms within these aquatic systems are affected by pesticide pollution. While many studies have examined the effects of pesticides on individual organisms, in a series of experiments, I investigated the effects of a pesticide mixture on members of a complex host-parasite system and on host susceptibility to infection. In my first experiment in the laboratory, I examined changes in survivorship when trematode parasites (Echinostoma trivolvis) and their first intermediate host, Planorbella trivolvis snails, were exposed to a low concentration (10 ppb: 15 ppb) and high concentration (85 ppb: 100 ppb) mixture of atrazine and metolachlor, respectively. There was a significant decline in parasite survivorship in the high concentration treatment at 14 hours, while snail survivorship was unaffected across all treatments. In my second experiment, prior to infection, I exposed the parasites and/or second intermediate hosts, Rana clamitans and Rana sylvatica tadpoles, to the pesticide mixtures and examined subsequent infection levels in the tadpoles. The atrazine and metolachlor mixtures had no significant effects on parasite load in the laboratory. Newly shed parasites were more likely than 10 hours old parasites to infect tadpoles, regardless of pesticide exposure. In my final experiment, I utilized outdoor mesocosms to expose parasites, snail hosts, and Rana sylvatica tadpoles to the pesticide mixture, and I examined differences in parasite load within the tadpoles after two weeks. The pesticides had no significant effect on parasite loads in the field. Overall, my findings suggest the atrazine and metolachlor mixtures used in this study had no significant effects on disease dynamics in a system involving Echinostome parasites, snails, and tadpoles. Master of Science