Following deglaciation, Canadian coastlines, particularly those around Hudson and James Bay region were affected by large crustal rebound and associated sea level change, the largest glacio-isostatic recovery process known during the Holocene. Recent archaeological work in the uplifting eastern James Bay region found unusual occupation and settlement patterns for early indigenous communities which seem to have been related to changes in the coastal landscape. Although numerous studies were conducted in the region there are several knowledge gaps and methodological limitations that render unclear the patterns and mechanisms of coastal landscape change in the region and, in general, on rebounding coastlines. For instance, the chronological data for sea level changes in northern Canada is largely based on radiocarbon dating of marine shells, which are inherently problematic due to the marine reservoir and the isotopic fractionation effects. Also, there is limited knowledge on forest and wetland ecosystem development as well as fire disturbance processes along the uplifting eastern James Bay shoreline. Finally, there is virtually no understanding of the mechanisms involved in changes of coastal resources availability for indigenous communities despite recent concerns about resource scarcity and change.My research attempts to fill in the aforementioned knowledge gaps and clarify the mechanisms involved in the postglacial coastal change under the influence of glacio-isostatic uplift in eastern James Bay. My research began with a revaluation of the large-scale shoreline displacement processes and proposes a new sea level change model based on improved radiocarbon methodology and recognition of ecosystem change. My results show that during the last 7 ka the shoreline elevation has regressed at a decreasing rate. The rate of shoreline emergence was initially rapid (6.5 m/100 yr) between 6850 and 6400 cal yr BP then slowed down to 1.4-2 m/100 yr during the late Holocene. In addition, this study relates the shoreline replacement patterns in the region with patterns of human occupation and offers an explanation for the unusual archaeological finds. My subsequent research presents a thorough investigation of ecological changes along a 100-km long and 6,500 years old chronosequence in the form of two studies focused on the main regional ecosystems – the boreal forest, including fire processes as the main landscape disturbance in the region, and wetlands. The former study documents increased fire activity near archaeological sites and suggest that Shield Archaic populations could have played a greater role in shaping their local landscape than previously thought. The regional forest dynamic patterns show a common developmental pathway, with alder thickets leading the colonization front onto the rebounding coastline followed by the typical boreal forest constituent – spruce. A steady increase in Pinus pollen, particularly after 800 cal yr BP, suggests a gradual immigration of fire-adapted Pinus from the south, probably in response to increasing fire activity. A decrease in tree pollen during the last 800 years indicates an opening of the taiga. The latter study brings empirical evidence of complex wetland ecosystem succession and uses this to test established theories of boreal wetland succession. My finds show that wetland succession on rebounding coastlines does not follow the classical model of hydrarch succession and suggests that succession does not imply the existence of a climax state but simply a sequential replacement of vegetation communities. The final component of my thesis evaluates the relationship between landscape change and availability of coastal resources, being driven by recent concerns about changes in coastal resource availability for the aboriginal communities. / Après la déglaciation, les littoraux canadiens, particulièrement ceux des région de la Baie James ont été affectés par d'importants mouvements de la croûte terrestre et par les changements du niveau de la mer y associés. De récentes études archéologiques effectuées dans la région à l'est de la Baie James ont prouvé l'existence de peuplements et d'occupations inhabituels qui semblent avoir été liés à des changements dans le paysage côtier. Bien que plusieurs études aient été menées dans la région, il existe plusieurs lacunes dans les connaissances et des limitations méthodologiques qui complexifient la compréhension des mécanismes du changement du paysage côtier de la région. En outre, les connaissances sur le développement des écosystèmes forestiers et des zones humides ainsi que sur les processus de perturbations causée par des feux de foret a travers du Baie James sont limitées. Finalement, on ne comprend que très peu les mécanismes impliqués dans l'évolution de la disponibilité des ressources côtières pour les communautés autochtones, et ce, en dépit des préoccupations récentes concernant la pénurie et les modifications des ressources.Ma recherche vise à combler les lacunes dans les connaissances précédemment mentionnées et clarifier les mécanismes impliqués dans la modification postglaciaire côtière influencée par de relèvement glacio-isostatiques dans l'est de la Baie James. Ma recherche a commencé par une réévaluation des processus de déplacement du littoral et propose un nouveau modèle du changement du niveau de la mer fondée sur une méthodologie 14C améliorée et la reconnaissance de l'évolution des écosystèmes. Mes résultats montrent que durant les 7 derniers ka, l'élévation du littoral a reculé à un taux décroissant. Le taux d'émergence du littoral a initialement été rapide (6,5 m/100 ans) entre 6850 et 6400 cal BP, puis a ralenti à 1,4 à 2 m/100 ans à la fin de l'holocène. De plus, cette étude porte sur les modèles de remplacement du littoral dans la région, en tenant compte de l'occupation humaine, et propose une explication pour les découvertes archéologiques inhabituelles. Mes recherches subséquentes présentent une investigation sur les changements écologiques le long d'une chronoséquence ayant une longueur de 100 km et étant âgée de 6500 ans. Cette investigation se traduit en deux études portant sur les principaux écosystèmes régionaux - la forêt boréale, comprenant la dynamique des feux de foret en tant que perturbation principale du paysage, et les zones humides.L'étude documente le nombre accru d'incendies à proximité des sites archéologiques et suggère que les populations archaïques du Bouclier Canadien pourraient avoir joué un rôle plus important qu'on ne le pensait dans l'élaboration de leur paysage local. Les modèles de la dynamique forestière régionale montrent une voie de développement commune, avec d'aulnes favorisant la colonisation sur la côte relevées, suivi par la forêt boréale – l'épinette. Une augmentation du pollen de pin, particulièrement après 800 ans cal BP, suggère une immigration progressive du sud du pin, probablement en réponse à l'augmentation de la fréquence des incendies. Une diminution du pollen des arbres apres 800 BP indique une ouverture de la taïga.La dernière étude presente la succession de terre humides et les utilise afin de tester les théories établies à propos de la succession des zones humides boréales. Mes découvertes montrent que la succession des terres humides sur les littoraux en relèvement ne suit pas le modèle classique de la succession écologique et suggère que la succession n'implique pas l'existence d'un état climax, mais tout simplement un remplacement séquentiel des communautés végétales.Le dernier volet de ma thèse évalue la relation entre l'évolution du paysage et la disponibilité des ressources côtières, étant motivés par des préoccupations récentes concernant des changements dans la disponibilité de ces ressources pour les communautés autochtones.

North of Grand Rapids, Manitoba, new exposures of a carbonate succession prompted study of the lithology and paleontology of the uppermost Ordovician along the northern edge of the Williston Basin in Manitoba. Modern concepts and approaches were applied in examining the sedimentary rocks and fossils, including X-ray diffraction, stable carbon and oxygen isotope analysis, and statistical algorithms. Nine lithofacies, representing a series of shallowing events, and environmentally significant subaerial exposure surfaces, were identified. The distribution and relative abundance of identified fossils were used to delineate faunal associations, which were examined in the context of the impending end-Ordovician mass extinction. Historically, the stratigraphy of the latest Ordovician has been problematic. Therefore, detailed examination of this succession aided in clarifying unit boundaries in the Stony Mountain and Stonewall formations. Study of this new succession contributed a wealth of information to the understanding of the uppermost Ordovician of Manitoba.

The fossil-rich deposits of the latest Maastrichtian Frenchman Formation (66Ma) of Saskatchewan, Canada, contain a complete, continuous record of terrestrial vertebrate diversity in the last half-million years prior to the end-Cretaceous mass extinction. The objective of these studies was to recover biodiversity patterns at the level of the ecological community, and to explore the paleoenvironmental and paleoclimatic variables that influenced these patterns. The primary study locality for this research was the East Block of Grasslands National Park (GNP) in southern Saskatchewan. The first chapter of this thesis is review of paleoecology: what it is, why it is important, and how it is studied. The second chapter is a survey of the depositional environments in the Frenchman Formation in GNP, based on 23 mapped stratigraphic sections, as well as on sediment and paleosol (ancient soil) samples. Stratigraphic surveys revealed a series of repeated depositional cycles, each likely linked to an uplift and subsistence event. Based on these cycles, the Frenchman Formation was divided into seven lithostratigraphic units, across which paleoecological and paleoenvironmental trends could be more easily assessed. Paleoclimate data was determined using x-ray florescence (XRF) analysis of paleosol samples collected from two of the stratigraphic sections. No significant temporal trends in mean annual temperature or precipitation were recovered, although other sedimentological indicators imply that small-scale or seasonal climate fluctuations did occur. In Chapter 3, a second set of paleoclimate estimates is presented, with these estimates derived from plant macrofossil (leaf) physiogamy found in two Frenchman Formation paleofloral sites. Climate estimates were calculated using leaf margin analysis (LMA) and the Climate Leaf Analysis Multivariate Program (CLAMP), and compared with other paleoclimate estimates for the Western Interior Basin. The LMA and CLAMP estimates were not significantly different from those derived from the paleosol analyses, and were broadly similar to estimates from coeval sites in Montana and Wyoming. Based on the presence of inertinite (fossil charcoal) in one paleofloral site, the marked disparity in floral diversity between the two sites was suggested to be reflective of ecological succession following a disturbance by fire. Chapter 4 explores trends in vertebrate diversity throughout the Frenchman Formation, based on some 7400 fossils from 38 vertebrate microsites. Fossils were collected, along with the lithostratigraphic information necessary to place them in a chronological context, and were sorted, identified and catalogued. Temporal variations in alpha (within-site) and beta (among-site) diversities were explored. On the level of the ecological community, no significant overall trends in vertebrate diversity were recovered, although small-scale biodiversity patterns were discovered. A distinct turnover in turtle family diversity was found, coinciding with an increased occurrence of coastal-like taxa in the upper part of the formation. These diversity trends may be linked to increased brackish-water conditions associated with the Cantapeta Advance, a marine transgression described from coeval deposits in Wyoming. This study lends support to the catastrophic theory of the end-Cretaceous extinction, and underlines the importance of exploring both local and regional driver of paleodiversity. Furthermore, it demonstrates that focusing on the pre-extinction diversity patterns of single vertebrate groups (i.e. dinosaurs) can be misleading, as the group diversity can fluctuate independently of the overall diversity. These results have important implications for our understanding of the timing and cause of the Cretaceous mass extinction on local scales, and demonstrate the benefits of assessing paleodiversity patterns on multiple ecological levels. / Les dépôts du Maastrichtien (66Ma) de la formation Frenchman (Saskatchewan, Canada) sont connus pour leur richesse fossilifères. Ils contiennent ainsi un enregistrement continu et complet de la diversité des vertébrés terrestres pour les dernières 500 milles ans avant l'extinction de masse de la fin du Crétacé. L'objective de ces études était de reconstruire les patrons de la biodiversité au niveau de l'écologie des communautés, et d'explorer les variables paléoenvironnementales et paléoclimatiques qui ont influencé ces patrons. La principale localité étudiée était le Bloc Est du Parc National des Prairies dans le sud du Saskatchewan. Le deuxième chapitre se concentre sur l'étude des environnements de dépôts de la formation Frenchmann (GNP), reposant sur l'analyse de 23 sections stratigraphiques cartographiées, ainsi que sur des échantillonnages sédimentaires et des paléosols (sols anciens). L'étude stratigraphique a révélé une série de cycles de dépôts répétés, et en s'appuyant sur ces cycles, la formation Frenchman peut alors être divisée en 7 unités lithostratigraphiques. Les données paléoclimatiques ont été déterminées grâce à une analyse par spectrométrie de fluorescence X (SFX) des échantillons de paléosols. Aucune tendance temporelle significative dans la moyenne annuelle des température ou des précipitations n'a pu être enregistrée, bien que d'autres indicateurs sédimentaires montrent que le climat a pu fluctuer de façon saisonnière ou á une très petite échelle. Le chapitre 3 présente une deuxième ensemble de d'estimations paléoclimatiques dérivées provenant de la physiogamie de macrofossile de plantes (feuilles) trouvés dans les sites paléofloraux des deux formations Frenchman. Les estimations climatiques ont été calculées grâce à l'analyse des bords des feuilles ainsi que le Climate Leaf Analysis Multivariate Program (CLAMP). Les estimations basées sur les paléoflores ne présentaient pas de différences significatives avec celles obtenues par l'analyse des paléosols, et étaient majoritairement semblables aux estimations des sites du même âge dans le Montana et le Wyoming. Reposant sur la présence d'inertinite (charbon de bois fossile) dans un des sites paléofloraux, l'importante disparité observée au sein de la diversité florale des deux sites a été interprétée comme reflétant la succession écologique faisant suite à des feux de forêts. Le chapitre IV explore les tendances de la diversité au sein des assemblages de vertébrés au sein de la formation Frenchman. Pour se faire ont été étudiés environ 7400 fossiles récoltés sur les 38 microsites à vertébrés du Parc National des Prairies. Ils ont ensuite été triés, identifiés et catalogués. S'en est suivie une exploration des variations temporelles des diversités alpha et beta. D'un point de vue de l'écologie des communautés, aucune tendance significative au sein de la diversité des vertébrés n'a été observée, même si les diversités des groupes indépendants de vertébrés ont effectivement varié, généralement de façon asynchrone. Un renouvellement distinct de taxons de tortues a été noté, coïncident avec une occurrence croissante de taxons côtiers dans la partie supérieure de la formation. Cette étude démontre que le déclin de la diversité pré-extinction, annonciateur de l'extinction de masse caractéristique du Crétacé, n'était pas global dans la nature. De plus, elle démontre que se concentrer sur un seul groupe de vertébrés (i.e. dinosaures) pour comprendre les patrons de la diversité pré-extinction peut conduire à des interprétations trop générales, la diversité d'un groupe pouvant fluctuer indépendamment au sein de la diversité totale. Ces résultats ont d'importantes implications pour notre compréhension de la chronologie et des causes de l'extinction de masse Crétacé sur des échelles locales, et démontrent les bénéfices d'évaluer les patrons de la paléodiversité sur des échelles écologiques plus larges.

North of Grand Rapids, Manitoba, new exposures of a carbonate succession prompted study of the lithology and paleontology of the uppermost Ordovician along the northern edge of the Williston Basin in Manitoba. Modern concepts and approaches were applied in examining the sedimentary rocks and fossils, including X-ray diffraction, stable carbon and oxygen isotope analysis, and statistical algorithms. Nine lithofacies, representing a series of shallowing events, and environmentally significant subaerial exposure surfaces, were identified. The distribution and relative abundance of identified fossils were used to delineate faunal associations, which were examined in the context of the impending end-Ordovician mass extinction. Historically, the stratigraphy of the latest Ordovician has been problematic. Therefore, detailed examination of this succession aided in clarifying unit boundaries in the Stony Mountain and Stonewall formations. Study of this new succession contributed a wealth of information to the understanding of the uppermost Ordovician of Manitoba.

<p> The current biodiversity crisis is challenging the ability of conservation biologists to both monitor ongoing declines and create effective management plans. Ongoing habitat destruction, pollution, introduction of invasive species, and the initial stages of climate change are only some of the anthropogenic stresses that face today's biosphere. On human time scales, these changes are unprecedented, curtailing the availability of knowledge regarding ecological responses to stress and disturbance. The fossil record provides numerous disturbances of varying magnitudes throughout the history of life, and yet this resource has been often overlooked or dismissed by biologists. The purpose of this study is to examine the effects of stress on communities using methods that allow integration of modern and fossil data. With this goal in mind, various levels of disturbance are investigated across increasing temporal and spatial scales.</p><p> At the smallest spatial and temporal scale, I examined the effect of recent lake acidification on plankton communities, using techniques commonly applied by ecologists, as well as introducing a new method based on a well-established technique. Throughout this thesis, I use the Buzas-Gibson evenness metric and Non-metric Multi-Dimensional Scaling analysis (an ordination technique), as well as applying Rank-Abundance Curve Kurtosis, which measures the shape of species-abundance distributions. Each of these techniques is a different way of representing community structure, with each metric providing slightly different information. Within the lake acidification system, all communities displayed a shift in community structure as pH dropped, and again when pH values returned to neutral, indicating a gradual recovery from acid stress. The timing of this change reveals the ability of different communities to resist acidification, and the resilience of those communities through the recovery phase.</p><p> To determine the feasibility of comparing modern and fossil data, I selected four unrelated datasets with distinct disturbance events to represent different time scales, from two decades to one million years. Each dataset displayed a similar pattern; the disturbance event created a distinct shift in community structure followed by a gradual recovery after the stress levels decreased. A major concern when comparing modern and fossil data is the difference in temporal resolution, and specifically the effect of time-averaging which is expected to obscure ecological signals. Instead, I found that applying a model of time-averaging across the community data reduced background noise, thereby clarifying the pattern of ecological change observed in the raw data.</p><p> Extending the temporal and spatial scale, I explored the ecological response of marine microfossil assemblages during three intervals of rapid global warming, as analogues for modern global warming. Four taxonomic groups were included, two benthic and two planktic. Overall, diversity within communities increased during global warming, however this was due to the response of the planktic groups, as both benthic groups showed decreases in diversity. These findings support the utility of the fossil record in examining past disturbances, by providing a useful prediction for biotic responses to global warming.</p><p> Representing the largest spatial and temporal scale is the Botomian mass extinction (mid-Early Cambrian). This mass extinction is the first recognized mass extinction in the history of life, and occurred during an interval of rapid evolution and faunal turnover. During the extinction interval, there was a distinct change in community structure and an associated increase in instability.</p><p> The findings of this study are unique; community structure displays a similar response to stress across various taxonomic groups, in different environments, and at multiple temporal scales. The commonality in community response to stress likely represents a fundamental feature of disturbed ecosystems. Not only is the comparison of modern and fossil data possible, such comparisons offer new discoveries relevant to conservation biology and about the very nature of life on Earth.</p>

The Fairholme Carbonate Complex is the southernmost and largest in an extensive stromatoporoid-dominated reef domain that developed in Alberta during Frasnian time. Its western margin is defined, and exhibits a major re-entrant 25 km wide named the Shark Embayment. A Devonian paleoclimatic model suggests that the western margin was subjected to seasonal oceanic upwelling of nutrient-rich and oxygen-poor waters that influenced the development of the carbonate buildups. / Lithostratigraphic units of the Fairholme Group are characterized in terms of a Rocky Mountain (RM) Conodont Zonation. In the western part of the study area, the Flume carbonate platform began to onlap the westernmost slope of the West Alberta Ridge (WAR) as early as latest Givetian time (norrisi Zone). To the east, the Flume platform did not onlap and begin to cover the crest of the WAR until the succeeding RM Zone 1. The overlying Upper Cairn Member (containing a unique coral biofacies) and Perdrix Formation range from the uppermost RM Zone 1 to Zone 4b. The Peechee Member is difficult to date, but is probably within Zones 5a to 5b; the overlying Grotto Member is no younger than Zone 5b. The Arcs Member is within Faunal Intervals (F.I.) 6-7, and the Ronde Member is within Faunal Interval 8 which includes the Frasnian-Famennian boundary near its top. The Mount Hawk ranges from Zone 4b/5a to F.I. 8. / Stromatoporoids were widespread and abundant calcareous benthos living in shallow, tropical, oligotrophic, and agitated marine environments. Thirty-two species representing 5 different orders of stromatoporoids are recognized in this study; thirty-one species comprise a diverse fauna in the Cairn Formation. Five stromatoporoid assemblages are defined in the Flume and Upper Cairn succession, and are correlated with RM conodont zones. Stromatoporoids exhibit mostly domical, but also bulbous, laminar, and dendroid growth forms that were genetically prescribed and only slightly modified by environmental factors. Trace element geochemistry of Devonian stromatoporoids indicates that these sponges originally secreted a calcitic skeleton, in contrast to Ordovician stromatoporoids which secreted an aragonitic skeleton. Evidence suggest that Paleozoic stromatoporoids eventually developed into mixotrophic organisms enabling them to construct large carbonate buildups, but they were also vulnerable to the devastating effects of paleoceanic upwelling of nutrient-rich waters.

Palaeomacroecology is the study of large scale patterns of species diversity in the fossil record, encompassing a variety of subtopics. This thesis also addresses a variety of these subtopics, making it difficult to define under one heading.The first portion of the thesis deals with a new package of software tools for the analysis of large scale datasets, with a specific focus towards palaeoecology and palaeogeography. These software tools have been combined into a package called fossil that has been released on the Comprehensive R Archive Network (CRAN), and is already being used by other palaeoecologists. While the majority of these tools had a basis in previous statistical methods, I have also independently developed a clustering algorithm for use with biogeographic datasets. This clustering algorithm is relational, non-Euclidean and non-hierarchical and as such is called Non-Euclidean Relational Clustering (NERC). NERC eliminates several of the assumptions common to most other clustering methods that are often violated by biogeographic data.The next portion of my thesis describes a new Triassic aged flora from Axel Heiberg Island in Nunavut. Macroecological studies typically use large databases compiled from individual samples; therefore, these individual samples represent the foundation on which macroecological analyses rest, and collection and description of new fossil bearing sites is vital to the advancement of palaeomacroecology.Chapter 5 is an analysis of the provinciality and beta diversity of dinosaurs in the Late Cretaceous of North America. This analysis found that contrary to previous studies, dinosaur genera were widespread across the continent and not restricted to small geographic ranges. Chapter 6 is the final culmination of my thesis, and where I see palaeomacroecology headed in the future. It is an analysis of how latitudinal diversity gradients in plants have changed through time. The analysis assesses the impact of changing climate in creating and sustaining the latitudinal diversity gradient, and lends support to the idea that temperatures are important drivers of the gradient.The final chapter is a summary of where palaeomacroecology has been, and where its future work might be best focused. While the field of palaeontology is vital to our understanding of large scale, especially temporally, patterns of species diversity, the field of palaeontology has an opportunity to advance our understanding at an even more rapid pace provided we ask the appropriate questions of our data. / La palaeomacroecology est l'etude des modeles a grande echelle de la diversite des especes dans les archives fossiles, et inclue une variete de sous-themes. Cette these adresse aussi une variete de ces sous-themes, ce qui en fait diffucult de definir sous une seule rubrique. La premiere partie de la these discute d'un nouvel ensemble d'outils logiciels pour l'analyse des ensembles de donnees a grande echelle, avec une attention particuliere a la paleoecologie et la paleogeographie. Ces outils logiciels ont ete combines dans un paquet appele fossil qui a ete publie sur le reseau Comprehensive R Archive Network (CRAN), et est deja utilise par d'autres palaeoecologists. Bien que la majorite de ces outils avait une base en preious methodes statistiques, j'ai aussi developpe independamment un algorithm de regroupement pour une utilisation avec des bases de donnees biogeographiques. Cet algorithme de regroupement est relationnelle, non-euclidienne et non-hierarchique et en tant que telle est appele Non-Euclidean Relational Clustering (NERC). NERC elimine plusieurs des hypotheses communes a la plupart des autres methodes de classification, et qui sont souvent violees par des donnees biogeographiques.La partie suivante de ma these decrit une nouvelle flore du Trias a l'ile Axel Heiberg, au Nunavut. Les etudes macroecologiques utilisent generalement de grandes bases de donnees compilees a partir des echantillons individuels et, par consequent, ces echantillons individuels representent le fondement del'analyse macroecologique, et la collecte et la description des nouveaux sites fossiliferes est indispensable a l'avancement de la palaeomacroecologie.Le Chapitre 5 est une analyse du provincialisme et de la diversite beta des dinosaures aux Cretace superieur en Amerique du Nord. Contrairement aux etudes precedentes, cette analyse a revele que les genres de dinosaures ont ete beaucoup plus repondus a travers le continent et ne se limitement pas a de petites aires geographiques. Le Chapitre 6 est l'aboutissement final de ma these, ou je vois dans quelle direction se dirige a la palaeomacroecologie. Il s'agit d'une analyse de la facon dont les gradients de diversite des plantes ont change au fil du temps. L'analyse evalue le role des changements climatiques dans la creation et le maintien du gradient latitudinal de diversite, et soutient l'idee que les temperatures sont d'importants moteurs de ce gradient.Le dernier chapitre resume l'evolution palaeomacroecologie dans quelle direction les travaux futurs devraient etre orientes. Bien que le domaine de la paleontologie sait vital pour notre comprehension des modeles de la diversite des especes a grande echelle, en particulier celle temporelle, le domaine de la paleontologie a une occasion de faire progresser notre comprehension a un rythme encore plus rapide, a condition de poser les bonnes questions.

Thesis (Ph.D.)--University of Nebraska-Lincoln, 2009. / Title from title screen (site viewed July 21, 2009). PDF text: xiv, 216 p. : col. ill., col maps ; 8 Mb. UMI publication number: AAT 3350259. Includes bibliographical references. Also available in microfilm and microfiche formats.

Thesis (Ph. D.)--West Virginia University, 2005. / Title from document title page. Document formatted into pages; contains ix, 163 p. : ill. (some col.), maps. Vita. Includes abstract. Includes bibliographical references (p. 150-160).

Thesis (Ph. D.)--University of Oregon, 2004. / Typescript. Includes vita and abstract. Includes bibliographical references (leaves 394-419). Also available for download via the World Wide Web; free to University of Oregon users.