This is to certify that I have examined this copy of a master's thesis by Laura J. Six and have found that it is complete and satisfactory in all respects, and that any and all revisions required by the final examining committee have been made. substrate affinities, I also tested whether rooting substrate affected allocation to aboveand below-ground structures, morphological traits, or overall performance (total biomass and plant nitrogen), consistent with differences in moisture availability between substrates.I tested whether understory species showed significant associations with CWD or the forest floor using 1200, 1 x 1 m quadrats in which I compared cover of CWD and forest floor (as measures of substrate availability) and the density of shoots of each species emerging from each substrate. To examine relationships between rooting substrate and plant performance, I excavated three common understory species: Tiarella trifoliata (preference for forest floor), Vaccinium parvifolium (preference for CWD), andMaianthemum dilatatum (no preference for either substrate). For each species, the below-ground structures (rhizomes and roots) and aerial shoots (stems and leaves) of 30 pairs of plants (one rooted in CWD, one in the forest floor) were measured, then dried to determine mass. Replicate samples of each substrate were collected every 3 wk (June through mid-September) to test for differences in volumetric moisture.Most species (83%) displayed a preference for substrate; of these, 75% preferred the forest floor. Species preferences for CWD or forest floor did not appear to be related to plant stature or any other morphological or life-history trait. Most species may be associated with the forest floor because it is a more predictable and stable substrate.Alternatively, preference for the forest floor may relate to competition with mosses: moss cover was more than twice as high on CWD as on the forest floor (87 vs. 38%). Patterns of biomass allocation in Maianthemum, Tiarella, and Vaccinium were consistent with differences in moisture availability between substrates: plants rooted in the forest floor (which was consistently drier over the growing season) allocated greater biomass to below-ground structures. However, effects of substrate on above-ground traits (e.g., leaf density, leaf area, and shoot height) were non-significant, and effects on below-ground traits were largely inconsistent with differences in moisture availability between substrates.Depth and lateral spread of rhizome systems may be determined by the physical structure of decaying logs rather than by resource availability. In addition, light may be so limiting in these forests that it masks the potential for species to respond to substraterelated differences in resource availability. Finally, it is possible that associations with substrates are driven not by differences in performance of established plants, but by differences in dispersal, germination, and/or early survival.