The "old-growth controversy" in the Pacific Northwest recognized thinning as the primary silvicultural practice for land managers to produce wildlife habitat while continuing to produce timber. For the foreseeable future, forest stands will be harvested to produce forest gaps and a patchwork of trees of different ages. In order to evaluate the effect of thinning on biodiversity, nine 15-year-old harvests of this type (age=70 years) were paired with adjacent old-growth and even-aged, unthinned "pole" stands. Since soil is the crucible of terrestrial biodiversity, it is critical to contrast the effects upon the forest floor of this future practice with current management. Soil and litter fauna were monitored along 250 meter transects (pitfall and Berlese sampling at ten-meter intervals) to meet the following objectives: 1) to determine biological diversity on public lands, per federal mandate, and use diversity as a management tool; 2) compare levels of biodiversity between three management strategies; 3) determine best methods to assay arthropod diversity; and 4) identify structural and environmental determinants of arthropod diversity and abundance. The study's hypotheses were: 1) old-growth forests will have greater arthropod diversity than thinned stands; 2) thinned stands will have greater arthropod diversity than unthinned stands; and 3) species found within old-growth stands, but not within unthinned pole stands, will also be found in thinned stands. This study contrasted nine Western Hemlock/Douglas-fir sites each with contrasting old-growth, thinned and unthinned pole management stands. Sites were equally blocked in Southern Oregon, the Coast Range, and the Cascade Mountains. No segment of the arthropod fauna. (i.e., pitfall-trapped epigeic macroarthropods, Berlese-extracted litter-dwelling meso- and microarthropods, or soil-dwelling microarthropods) exhibited a management (treatment) effect throughout the entire region. When the regional blocking was removed, within-region analysis generally revealed that old-growth was most distinct. Old-growth stands had the highest abundance of individuals, but were comprised of the fewest species. Thinned stands were characterized by the highest species richness. Within-region analysis revealed an interaction of management effects and specific locale effects; locale effects dominated for soil microarthropods and epigeic macroarthropods, while management options dominated for litter arthropods. Within the Southern Oregon region, I attempted to correlate arthropod community structure (canonical correspondence analysis (CCA) of within-stand samples) with a suite of soil chemical and microbiological descriptors. Full analysis of twelve variables within one exemplary stand revealed several potential trends (negative: dissolved organic carbon, soil moisture, distance from the beginning of the transect; positive: total CO��� field respiration, mineralizable nitrogen, water-induced respiration, substrate-induced respiration). Relatively shallow slopes and very low r-value coefficients of correlation characterized all statistical tests. Few of the trends apparent at one site were paralleled at more than one other site; at all sites potential correlates had very low r-values. No community revealed separate clouds in CCA analysis, indicating distinct "micro-communities" of arthropods inhabiting distinct micro-habitats. Lack of distinctive species assemblages and lack of correlation with microhabitat variables indicated that arthropods respond on different temporal and/or spatial scales then the microbial-oriented variables, and that each taxon is responding in an individual manner. / Graduation date: 1998