The hemlock woolly adelgid was discovered in the southern Appalachians as early as 2002. It has caused widespread mortality of eastern hemlocks in the northern Appalachians and researchers have predicted a similar effect in the southern Appalachians. This study characterizes the early-stage impacts of hemlock woolly adelgid on eastern hemlock forests throughout the southern Appalachians. Twenty-eight permanent plots were reinventoried in 2004 and 2005. They were originally inventoried between 1990 and 1998 before the introduction of the hemlock woolly adelgid. The plots were located in Great Smoky Mountains National Park, Joyce Kilmer Memorial Forest, and Ellicott Rock Wilderness Area in the southern Appalachians. They encompass a wide geographic range, a wide range of environmental variables, and different levels of eastern hemlock dominance. The plots were divided into either old-growth forests or second-growth forests based on visual inspection. Within each 1000 m2 plot, vegetation inventory was performed for both herbs and woody species. Changes in eastern hemlock abundance and changes in species richness between the two inventory dates were examined. Environmental variables were correlated with the changes to determine if the changes were consistent with those caused by the hemlock woolly adelgid. This work found that small < 2.5 cm dbh eastern hemlock trees decreased in density in both old-growth and second-growth plots when reinventoried after arrival of the hemlock woolly adelgid. The percentage change in eastern hemlock density was correlated with the initial eastern hemlock importance value in both categories. Change in eastern hemlock density was not correlated with observed environmental variables, location, or successional stage. Instead, this study found that successional stage was relevant to the change in species richness after the introduction of the hemlock woolly adelgid. Old-growth forests increased in species richness whereas second-growth forests decreased in species richness. Initial eastern hemlock basal area, pH, and elevation were correlated with species richness in old-growth forests and measures of eastern hemlock dominance were correlated with the change in species richness in old-growth forests. In second-growth forests, however, pH was the most important variable in determining species richness and only nitrogen was correlated with the change in species richness. The contrasting response of species richness in reinventoried old- and second-growth plots may be reconciled by examining the differences in initial eastern hemlock dominance in old-growth and second-growth forests. Eastern hemlock importance value varied based on stage, with old-growth forests having a higher importance value than second-growth forests. The different responses between old growth and second growth may be explained by the initial difference in eastern hemlock importance value. Eastern hemlock mortality has been swift and widespread. If the hemlock woolly adelgid cannot be controlled, eastern hemlock will likely decline precipitously in importance throughout the southern Appalachians.