This dissertation is concerned with the relationships of prehistoric lithic technology to past subsistence and settlement systems operative at upper elevations in the Southern Interior Plateau of British Columbia. It has both a methodological and a substantive aspect. From a methodological perspective, the research applies a linear model of chipped-stone tool manufacturing processes and multivariate data reduction techniques to a series of lithic assemblages from surface sites located in Upper Hat Creek Valley in order to study intersite variability of stone tool manufacture and use. In order to efficiently study differences in tool manufacturing sequences, potential attributes that may measure such technological variability were evaluated by a R-mode factor analysis of a small sample of the assemblages; This enabled the selection of a reduced number of attributes that measure the underlying patterns of relationships present in the sample. Tool classes based on overall morphology and working edge characteristics were employed to describe artifact use. Two site classifications, one based on lithic waste patterning and the other on tool assemblage variability, were established by cluster analysis and multidimensional scaling of the assemblages. The relative effectiveness of each site classification as a means of delineating settlement types was evaluated by the ability to interpret the results of the analyses with respect to such variables as the nature and intensity of occupation, and environmental relationships. Interpretations of each analysis tend to be in general agreement with each other, although some differences are present. In some cases, the debitage analysis provides a more detailed and complex perspective of the type of occupation.- Also, a larger amount of patterning with environmental variables is evident among the site groups based on technological variability. Nevertheless, interpretations of site utilization based on results of both analyses were much more comprehensive relative to those possible from the examination of each analysis separately. In addition to studying interassemblage variability, the analysis of debitage provided some insight into the quantitative patterning of technological attributes and their significance as measures of variation in manufacturing steps. In two specific instances, observed attribute patterning is opposite to that expected by present knowledge of lithic technology. The explanation of these differences indicates some directions to be pursued by future experimental studies. The empirical validity of each site classification was also investigated. A series of Kruskal-Wallis analysis of variance tests was run on general lithic assemblage data and technological attributes to determine if the site groups defined are statistically significant. The tool-based site classification differentiates general lithic assemblage variability better than debitage however, this tends to reflect site size rather than technological processes. Both analyses support expectations based on ethnography that Upper Hat Creek Valley was likely utilized for seasonal hunting and plant gathering. These activities are reflected by the two main settlement types defined: 1) staging sites for hunting and butchering activities and 2) plant gathering and processing sites. Considerable variation with respect to the emphasis on extractive and maintenance activities is present within each type. This study has major implications for the future study of interassemblage variability where the predominant artifact class is lithic debitage. It has demonstrated that technological patterning is observable at the intersite level and that this can be accounted for in terms of subsistence-settlement activities. / Arts, Faculty of / Anthropology, Department of / Graduate