E astern hemlock is a major component of riparian forests in the southern Appalachian Mountains (Narayanaraj et al., 2010). As the dominant conifer in riparian areas, this species plays an important role in regulating nutrient cycling processes and climatic conditions in both terrestrial and aquatic environments (Ellison et al., 2005). Th e hemlock woolly adelgid (Homoptera: Adelgidae), an exotic pest, was introduced to the mid-Atlantic region of North America from Asia in the 1950s and has spread throughout most of the range of eastern hemlock, moving north toward Canada and to the southern Appalachians. Pontius et al. (2006) examined the mortality of eastern hemlock following infestation with HWA and found that some trees can live for >10 yr following infestation. From county-level data on the spread of HWA, Evans and Gregoire (2007) found the rate of spread in Pennsylvania and north was 8.13 km yr −1 , while the rate of spread in the south was greater: 15.6 km yr −1 . Th e diff erence was attributed to colder temperatures in the north. Morin et al. (2009) modeled the spread of HWA north and south. While they concurred that spread to the north was limited by temperature, they concluded that the rate of spread to the south and west was related to host density. Indeed, in the southern part of hemlock's natural range, infestations were fi rst observed in the early 2000s and the rate of spread and mortality have been much faster than observed in northern areas Nuckolls et al., 2009 Eastern hemlock [Tsuga canadensis (L.) Carrière] trees serve an important ecological role in riparian ecosystems in the southern Appalachians. Signifi cant hemlock mortality is occurring due to infestation by the hemlock woolly adelgid (HWA) (Adelges tsugae Annand), a non-native invasive pest. Our objective was to quantify the impacts of HWA and hemlock mortality on nutrient cycling pools and processes. In 2004, we established eight research plots in riparian areas with >50% basal area in hemlock and four reference plots in riparian areas without hemlock (hardwood). All hemlock plots were infested with HWA. In four of the hemlock plots, all hemlock trees were girdled to induce defoliation and rapid mortality. By fall 2006, there was 90 and 10% mortality in the girdled and nongirdled hemlock plots, respectively. Measurements included soil temperature and moisture, nutrient pools, N transformations, litterfall and forest fl oor amount and chemistry, and throughfall and soil solution chemistry. From 2004 to 2008, litterfall composition changed, with an initial increase in the hemlock needle percentage followed by a decline. Hemlock plots had cooler spring soil temperatures than hardwood plots. Hemlock plots had greater surface soil and forest fl oor total C than hardwood plots; soil C content did not change during the 4 yr of measurement. Th ere were no diff erences in N mineralization rates or soil solution N concentrations among treatments. Diff erences between litterfall and forest fl oor nutrient contents in hemlock and hardwood plots ...