Title from PDF of title page (University of Missouri--Columbia, viewed on Feb 25, 2010). The entire thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file; a non-technical public abstract appears in the public.pdf file. Dissertation advisor: Dr. H. Carl Gerhardt. Vita. Includes bibliographical references.

Thesis (Ph. D.)--University of Texas at Austin, 2001. / Vita. Includes bibliographical references (leaves 87-99). Available also in a digital version from Dissertation Abstracts.

Thesis (Ph. D.)--University of Missouri-Columbia, 2000. / Typescript. Vita. Includes bibliographical references. Also available on the Internet.

Thesis (Ph. D.)--University of Missouri-Columbia, 2000. / Typescript. Vita. Includes bibliographical references. Also available on the Internet.

Understanding the nature of genetic variation in natural populations is an underlying theme of population genetics. In recent years population genetics has benefited from the incorporation of landscape and environmental data into pre-existing models of isolation by distance (IBD) to elucidate features influencing spatial genetic variation. Many of these landscape genetics studies have focused on populations separated by discrete barriers (e.g., mountain ridges) or species with specific habitat requirements (i.e., habitat specialists). One difficulty in using a landscape genetics approach for taxa with less stringent habitat requirements (i.e., generalists) is the lack of obvious barriers to gene flow and preference for specific habitats. My study attempts to fill this information gap to understand mechanisms underlying population subdivision in generalists, using the squirrel treefrog (Hyla squirella) and a system for classifying ‗terrestrial ecological systems‘ (i.e. habitat types). I evaluate this dataset with microsatellite markers and a recently introduced method based on ensemble learning (Random Forest) to identify whether spatial distance, habitat types, or both have influenced genetic connectivity among 20 H. squirella populations. Next, I hierarchically subset the populations included in the analysis based on (1) genetic assignment tests and (2) Mantel correlograms to determine the relative role of spatial distance in shaping landscape genetic patterns. Assignment tests show evidence of two genetic clusters that separate populations in Florida‘s panhandle (Western cluster) from those in peninsular Florida and southern Georgia (Eastern cluster). Mantel correlograms suggest a patch size of approximately 150 km. Landscape genetic analyses at all three spatial scales yielded improved model fit relative to isolation by distance when including habitat types. A hierarchical effect was identified whereby the importance of spatial distance (km) was the strongest predictor of patterns of genetic differentiation above the scale of the genetic patch. Below the genetic patch, spatial distance was still an explanatory variable but was only iv approximately 30% as relevant as mesic flatwoods or upland oak hammocks. Thus, it appears that habitat types largely influence patterns of population genetic connectivity at local scales but the signal of IBD becomes the dominant driver of regional connectivity. My results highlight some habitats as highly relevant to increased genetic connectivity at all spatial scales (e.g., upland oak hammocks) while others show no association (e.g., silviculture) or scale specific associations (e.g., pastures only at global scales). Given these results it appears that treating habitat as a binary metric (suitable/non-suitable) may be overly simplistic for generalist species in which gene flow probably occurs in a spectrum of habitat suitability. The overall pattern of spatial genetic and landscape genetic structure identified here provides insight into the evolutionary history and patterns of population connectivity for H. squirella and improves our understanding of the role of matrix composition for habitat generalists.

Sun-compass orientation tests were performed on adults and larvae of the spring peeper (Hyla crucifer) and the western chorus frog (Pseudacris triseriata triseriata). All animals were tested outdoors in a circular test pool from which all visible landmarks were obscured. Response on the Y-axis varied with the life history stage, and significant interspecific differences in sun-compass orientation behavior were noted. Statistical analysis was performed with Batschelet's V-test.Adult frogs were captured as they migrated to the breeding pond and tested under sunny skies. Chorus frogs oriented on the deep-water Y-axis of the nearest shoreline (n=30, p <0.005) but spring peepers did not orient (n=42, p >0.10).Frogs were also captured in the breeding pond at weekly intervals throughout the breeding season and tested under sunny skies. Chorus frogs oriented toward shore on the Y-axis (n=45, p <0.005); spring peepers did not orient (n=96, p >0.10). Adult chorus frogs kept in outdoor enclosures for a month after the breeding season were still oriented toward shore on the Y-axis (n=6, p <0.05). Chorus frogs tested under sunny skies after ten days of constant temperature, dim-light conditions in the laboratory did not orient (n=10, p >0.10).Adult chorus frogs captured as they left the breeding pond and tested under sunny skies did not orient on the Y-axis of the nearest shoreline (n=33, p >0.10), and neither did spring peepers (n=38, p >0.10).Larval western chorus frogs raised in outdoor enclosures oriented toward deep water on the Y-axis within five days after hatching (n=29, p< 0.005) and during late premetamorphosis (n=33, p <0.001). Chorus frog larvae did not orient during mid-premetamorphosis (n=32, p >0.10). Newly metamorphosed juveniles oriented on the Y-axis toward shore within a week after emergence (n=25, p< 0.01).Larval spring peepers oriented on the Y-axis toward shore within 18 days after hatching (n=37, p <0.0001) and also late in metamorphosis (n=29, p< 0.0001). Newly metamorphosed juveniles were not tested.

Thesis (Ph. D.)--University of Texas at Austin, 2007. / Vita. Includes bibliographical references.

Submitted by Alberto Vieira (martins_vieira@ibest.com.br) on 2018-06-15T20:06:11Z No. of bitstreams: 1 273713.pdf: 18588455 bytes, checksum: 1657ce57f9b07d6ecf7314dd1e71bcce (MD5) / Made available in DSpace on 2018-06-15T20:06:11Z (GMT). No. of bitstreams: 1 273713.pdf: 18588455 bytes, checksum: 1657ce57f9b07d6ecf7314dd1e71bcce (MD5) Previous issue date: 1995-08-18 / CAPES / O grupo de Hyla rubicundula é atualmente composto por três espécies: H. rubicundula Reinhardt & Lütken, 1862; H. tritaeniata Bokermann, 1965; H. anataliasiasi Bokermann, 1972. Soma-se a estas H. elongata Lutz, 1925, considerada sinônimo júnior de H. rubicundula. A ausência de traços marcantes diagnósticos entre estas espécies tem tornado difícil a identificação precisa das mesmas, fazendo-se necessário um estudo de caráter taxonômico. Para tal, reuniu-se material de diversas coleções, sobre o qual nos detivemos exclusivamente no exame da morfologia externa, através da caracterização de padrões de forma e desenho e de análises morfométricas multivariadas ( análise discriminante múltipla e análise dos componentes principais). O resultado inicial forneceu treze morfoespécies, para as quais foram analisadas as variações intra e interpopulacionais. Encontrou-se um alto grau de heteromorfismo dentro e entre tais amostras, o que dificultou o estabelecimento de limites específicos. Parte desta variação foi relacionada à variação ontogenética, sendo também em parte relacionada à variações individuais e regionais. Destas treze morfoespécies, nove foram consideradas espécies válidas. Hyla elongata teve sua sinonimização à H. rubicundula validada; contudo, faz-se necessário designar um lectótipo e paralectótipos para H. elongata, visto este táxon ter sido descrito com base em material colecionado em regiões diversas e ser composto por mais de uma espécie. / The Hyla rubicundula group is currently composed by three species: H. rubicundula Reinhardt & Lütken. 1862; H. tritaeniata Bokermann. 1965; H. anataliasiasi Bokermann, 1972. We can add to these H. elongata Lutz, 1925, considered junior synonym of H. rubicundula. The absence of strong diagnostic characters among these species makes difficult their accurate identification, making taxonomic studies necessary. With this objective, we have gathered material from several collections, in which we have studied exclusively the external morphology, by characterizing standards of form and pattern and through multivariate morphometrics analysis (multiple discriminant analysis and principal component analysis ). The first results provided thirteen morphospecies, for which were analyzed intra- and inter-populational variations. We have found a high degree of intra- and inter-species heteromorphism, which made difficult to establish limits among species. Part of these variation was associated to ontogenetic variation, and part was associated to individual and regional variations. Of these thirteen morphospecies, nine were considered valid species. Hyla elongata had its synonymization to H. rubicundula validated; however, it is necessary to designate a lectotype and paralectotypes for H. elongata, because the description of this taxon was based on material collected at diverse regions and composed probably by more than one species.

Submitted by Alberto Vieira (martins_vieira@ibest.com.br) on 2018-01-23T23:45:39Z No. of bitstreams: 1 493356.pdf: 35091289 bytes, checksum: 92371eb3bdd2b4b4fc6ceaad161b9ff8 (MD5) / Made available in DSpace on 2018-01-23T23:45:39Z (GMT). No. of bitstreams: 1 493356.pdf: 35091289 bytes, checksum: 92371eb3bdd2b4b4fc6ceaad161b9ff8 (MD5) Previous issue date: 2000-11-30 / CAPES / O grupo de Hyla circumdata é correntemente composto por onze espécies distribuídas em regiões serranas providas de córregos de montanha em ambientes de Floresta Atlântica, com altimetrias acima de 500 m. Este grupo é atualmente diagnosticado pela presença de faixas transversais na face posterior da coxa e pré-pólex hipertrofiado em machos adultos. Contudo, esta diagnose engloba outros grupos, como os de H. boans e de H. martinsi. A análise da morfologia externa de espécimes adultos, incluindo quinze caracteres morfométricos, girinos, vocalização, história natural e distribuição geográfica revelaram extrema variação intra e interpopulacional em alguns táxons do referido grupo, incluindo complexos gradientes de variação geográfica. Caracteres morfológicos, osteológicos e de história natural foram adicionados à diagnose do grupo em questão, objetivando-se eliminar sobreposições às diagnoses de grupos afins. Pela morfologia e hábitos, H. ibitiguara Cardoso, 1983 foi retirada do grupo de H. circumdata e agrupada com H. pseudopseudis Miranda-Ribeiro, 1937 e H. saxicola Bokermann, 1964. Hyla martinsi Bokermann, 1964 passa a constituir grupo à parte com H. langei Bokermann, 1965, caracterizado por apresentar espécimes com espinha umeral e pré-pólex bífido. Hyla clepsydra A. Lutz, 1925 e H. claresignata Lutz & Lutz, 1939 igualmente passam a compor grupo distinto, agrupados pela morfologia singular e girinos especializados a ambientes lóticos. Os grupos de H. pulchella e H. alvarengai foram mantidos independentes do grupo de H. circumdata por apresentarem, entre outras características... / The Hyla circumdata species group currently comprises eleven species, which are distributed over mountain stream habitats in Atlantic Tropical Forest, with altimetric range over 500 m. This group is diagnosed by transverse bands on the posterior surface of thighs and hypertrophied prepollex. However, this diagnosis comprises other groups, as H. boans and H. martinsi species groups. The analysis of external morphology of adult specimens, including fifteen morphometric characters, tadpoles, vocalization, natural history, and geographic distribution, revealed considerable intra- and interpopulation variation of some taxa, including geographic gradients of variation not easy to understand. Morphological characters, osteology, and natural history were added to the original diagnosis of H. circumdata species group, in order to avoid superpositions with allied species groups. Hyla ibitiguara was removed from the H. circumdata species group by general morphology and habits, and grouped with H. pseudopseudis Miranda-Ribeiro, 1937 and H. saxicola Bokermann, 1964. Hyla martinsi Bokermann, 1964 and H. langei Bokermann, 1965 were grouped together, characterized by developed humeral crest and bifid prepollex. Hyla clepsydra A. Lutz, 1925 and H. claresignata Lutz and Lutz, 1939 were grouped together, characterized by distinct external characters in adults and stream-adapted tadpoles. The H. pulchella and H. alvarengai species groups were not combined with H. circumdata species group, mainly by differences...

Male frogs typically call near water at dusk to attract females for breeding. During the breeding season, male treefrogs also emit diurnal "tree calls" or "rain calls" from the tops of trees. Very little is known about tree calls, although many treefrogs use them. Tree calls may be used to attract females, deter males or be triggered by weather conditions favorable for breeding: high temperature and relative humidity, and a drop in barometric pressure. As dusk approaches, male treefrogs continue tree calls from lower in the trees, and if conditions are favorable, jump to the ground and travel to a nearby breeding pond where they begin their repetitive nocturnal mating calls. The scant published information is mostly descriptive and does not address the fitness benefit of calling from treetops far from breeding ponds. My goal was to determine the function of tree calls based on their environmental, behavioral, and acoustic characteristics. My data indicate tree calls are not rain calls. Each treefrog species that I studied (Hyla femoralis, H. gratiosa, and H. squirella) called most frequently at different combinations of mean environmental characteristics (temperature, relative humidity, and barometric pressure). Hyla femoralis and H. gratiosa gave tree calls at ambient air temperatures that differed significantly from the distributions recorded when no treefrogs called. Temperature, relative humidity, and barometric pressure distributions of calling activity differed significantly among all three species and from the distributions recorded when no treefrogs called. Hyla squirella called most often at a significantly different mean relative humidity of 1015 mbar; whereas H. gratiosa and H. femoralis called at a median1017 mbar. Means and fluctuations (summarized as SD) of the three weather parameters explained significant variation in tree calling activity (32-60%). Tree calling activity for all three treefrog species were also not significantly affected by subsequent rain. These results indicate that tree calls were not given at random with respect to environmental conditions. My data suggest tree calls are advertisement calls that deter males from an area, as evidenced by partitioning of tree calls among species during the day. In a playback experiment conducted at Chuluota Wilderness Area, Florida (28o38.31'N 81o07.24'W) no significant effect on mating behavior was found for either call indicating that neither tree calls alone or in conjunction with mating calls are necessary for mating . However, due to habitat differences between treatments and a limited number of experimental replicates, further research is needed. Preliminary results indicate an additional four natural ponds should be sampled to determine a possible effect for tree calls. Acoustic analysis showed that tree calls had fewer pulses per call, more time between pulses within a single call, and a higher minimum call frequency than mating calls. Call duration and maximum call frequency of tree and mating calls did not differ significantly. My research has greatly increased the information known about tree calls. My results indicate tree calls are not only "rain calls," a common misperception about daytime tree calls. However, more research is needed to fully understand the function of tree calls.