xiv, 166 p. : ill. A print copy of this title is available through the UO Libraries. Search the library catalog for the location and call number. / Restoration of imperiled ecosystems has emerged as a national priority, but there is little mechanistic understanding of how to restore ecosystems so as to sustain both species diversity and ecosystem function. The main objectives of my dissertation were (i) to develop an understanding of mechanisms that structure upland and wetland prairie plant communities in Oregon's Willamette Valley, with particular focus on edaphic and competitive controls over native and exotic species, and (ii) to apply this knowledge toward more effective restoration of prairie ecosystems. I used a combination of experiments and analysis of natural gradients to examine the effects of succession, competition, and environmental heterogeneity on plant community structure and ecosystem function within a restoration framework. I conducted a large, replicated field experiment and a retroactive study of previously restored wetland prairies to assess different site preparation techniques. These techniques had variable effectiveness in suppressing the existing vegetation and seed bank, thus providing different initial successional trajectories. However, over time plant community structure converged due to a loss of early-successional species and the increasing dominance of native bunchgrasses; hence, there was a negative relationship between cover of native species and diversity. Only the more extreme treatments, such as topsoil removal, had large impacts on soil functioning. These studies underscore the importance of using a successional framework to guide restoration efforts. Given the potential importance of competition between native and exotic grasses in structuring prairie vegetation, I used a paired study of field and greenhouse experiments to determine how abiotic factors influence the competitive hierarchies between native and exotic grasses commonly found in upland and wetland prairies. Exotic grasses dominated competitive interactions with the native grasses, but this depended upon nutrient and moisture availability. Finally, I used a laboratory experiment to determine the seasonal and edaphic controls over nutrient and carbon cycling within a spatially heterogeneous upland prairie. Manipulating moisture and temperature resulted in significant changes in nitrogen, phosphorus, and carbon cycling, particularly in the winter. Under projected future climate change, these changes will likely have large effects on plant community structure. This dissertation includes my previously published and co-authored materials. / Advisers: Scott D. Bridgham, Barbara "Bitty" A. Roy, Bart R. Johnson