The acquisition of baseline data on Tsuga canadensis forests in the southern Appalachian Mountains has become increasingly important as these forests face imminent climatic changes and alterations induced by an exotic-invasive insect pest, the hemlock woolly adelgid (Adelges tsugae). While general forest patterns in the southern Appalachian Mountains are well documented, there is a dearth of research that specifically examines the characteristics of Tsuga canadensis forests in this region. To examine the compositional and environmental characteristics of these forests, I established 50 plots within mature Tsuga canadensis stands of the Great Smoky Mountains National Park (GSMNP), USA. I used cluster analysis to identify forest types; differences among types were assessed using analysis of variance (ANOVA) with Bonferroni multiple comparison tests. Detrended correspondence analysis (DCA) was used to relate Tsuga canadensis forest patterns to environmental variation. Four Tsuga canadensis forest types were identified in this study: Tsuga canadensis/Betula/Acer rubrum; Tsuga canadensis/Liriodendron tulipifera/Halesia carolina; Tsuga canadensis/Betula, and Tsuga canadensis. Site variables that were significantly different among forest types included slope aspect, litter depth, tree species richness, and diversity. DCA results separated Tsuga canadensis forests with northern hardwood affinities from those with cove hardwood affinities, as well as mesic from submesic forests. The DCA results also suggest gradients of canopy openness, soil properties, and location linked with the compositional variability among Tsuga canadensis forests of GSMNP. Ongoing forest dynamics and past anthropogenic disturbances prior to the formation of the national park are probable causal factors for the forest-environment linkages observed in this study.