Thesis (doctoral)--Grossherzogliche Rupprecht-Karls-Universität zu Heidelberg, 1902. / Vita. Includes bibliographical references (p. 71-75).

The distribution of herring (Clupea harengus pallasi) in the Northeast Pacific extends from southern California to northern Alaska. Studies on variation in herring life-history are limited to local characteristics and recruitment populations or relatively restricted regions of the Northeast Pacific. In this study I assessed herring life-history patterns and recruitment variation among 14 sites extending from Lynn Channel in Alaska to San Francisco Bay in California. Biological data were compiled from published and of state and provincial unpublished technical reports fisheries agencies in the northeast Pacific. Multivariate (PCA) and inferential statistical methods were applied in data analysis. Ordinations of length-and weight-at-age revealed no latitudinal patterns among the 14 herring sites. Among four sites for which environmental data were available, there were significant negative correlations between first PC scores of size and Ekman layer transport and sea-surface salinity (SSS). Reproductive characteristics of herring appeared to vary latitudinally. Herring from the more southerly sites tended to mature at an earlier age and smaller size and have a longer duration of spawning than herring from northerly sites. There were significant negative first PC scores of reproductive correlations between variables and Ekman transport, sea-surface temperature, and SSS. With the exception of Lynn and Seymour Channels in Alaska, the most northerly sites in this study, asymptotic to northern size (L[subscript]���) tended to increase from southern latitudes. With the exception of southern Strait of Georgia (British Columbia) herring and Tomales Bay (California) herring, growth coefficients (K) appeared to be higher in latitudes. L[subscript]��� was negatively correlated with SST. Recruitment variation at three sites was related to Ekman layer transport during the periods of spawning. At San Francisco Bay recruitment was negatively related to winter Ekman transport. At Sitka and Island, recruitment showed a southwestern Vancouver significant positive and negative correlation, respectively, Recruitment in northern and with spring Ekman transport. southern Strait of Georgia were negatively correlated with SST during fall. There was no correlation between recruitment and SSS for all sites. / Graduation date: 1997

Thesis (B.A.(Hons.)) -- Dept. of History, University of Adelaide, 1975.

Thesis (Ph. D.)--State University of New York at Buffalo, 1990. / Includes bibliographical references.

Stocks of salmon are declining in the Pacific Northwest. Based on region-wide studies that list and categorize the status of salmon stocks (Nehlsen et al., 1991; Huntington et al., 1994; and Nawa, 1995), I analyze the watersheds where stocks of salmon spawn for several anthropogenic variables, most of which are known to affect salmon. A total of 202 watersheds (stocks of salmon) in Washington, Oregon, California, and Idaho are coded for 13 anthropogenic variables such as dams, forest, agriculture, human population, hatcheries, Indian tribes, and watershed organizations. Most variables are ordinally coded 1 to 5 with the help of maps, some survey results, and visual assessment. Hypotheses are that the human activities that are detrimental to the health of salmon are associated with the poor status of salmon stocks in watersheds where they spawn. Salmon watersheds are also analyzed according to ecoregions and salmon management regions. Principal components analysis is performed to reduce the number of anthropogenic variables into factors. Kendall's tau, partial correlation, multiple linear regression, logistic regression, and discriminant analyses are performed separately for five species/races of salmon (coho, spring chinook, summer chinook, fall chinook, winter steelhead) and for the watershed average for salmon stock status, against the anthropogenic variables and factors. Statistical results using both variables and factors support most of the hypotheses relative to salmon and anthropogenic activities in watersheds. Results suggest that the 5 species/races of salmon are associated differently with most anthropogenic activities. The greater the amount of dams below watersheds where salmon spawn, the less healthy the salmon stocks. Summer chinook are least healthy where hatcheries are absent, while wild and scenic rivers are associated with healthier spring chinook. Coho are least healthy where there is more agriculture, more human population, and less forest. Fall chinook are least healthy where there is more US Forest Service land. Winter steelhead are least healthy where hatcheries are absent and less Indian tribal land. Ecoregional differences coupled with the absence of dams on the Coast compared to the Columbia Basin are associated with healthier coastal stocks. / Graduation date: 1998

Thesis (M.S.)--Oregon State University, 2009. / Printout. Includes bibliographical references (leaves 100-108). Also available on the World Wide Web.

Thesis (Ph. D.)--University of Washington, 1994. / Vita. Includes bibliographical references (leaves [240]-273).

Lupinus lepidus (Fabaceae) contains many morphologically divergent varieties and was restricted in its range during the last period of glaciation. A combination of phylogenetic (with the trnDT and LEGCYC1A loci) and population genetics approaches (with microsatellites and LEGCYC1A are used here to characterize intervarietal relationships and examine hypotheses of recolonization of areas in the Pacific Northwest affected by glaciation. Sequenced loci are not found to form a clade exclusive to L. lepidus, nor are any of the varieties found to form clades. Population genetics analyses reveal only negligible genetic structure within L. lepidus, with the majority of variation being found within populations. Isolation-by-distance analysis reveals some correlation between population genetic distances and geographic distance. Microsatellite and sequence results are consistent with a scenario whereby the Oregon and Washington regions were rapidly colonized from the south, with independent invasions along the eastern and western sides of the Cascade Mountains. A predicted disjunction between northern and southern populations is found within the microsatellite data but not the sequence data, suggesting that northern populations were recolonized via a process involving the spread of novel microsatellite mutations, perhaps through the persistence of a glacial refuge isolated from southern populations. Varieties are not shown to be genetically isolated, and are interpreted as representing ecotypes, with local selection outpacing the effects of migration.

Thesis (Ph. D.)--University of Washington, 1999. / Vita. Includes bibliographical references (leaves [98]-118).

No description available.