Eastern hemlock (Tsuga canadensis (L.) Carrière) is a foundation species in eastern North American forests, providing critical habitats for a number of species. These trees are experiencing widespread decline due to the spread of hemlock woolly adelgid (HWA: Adelges tsugae Annand Order Hemiptera) into their range, potentially resulting in the disappearance of hemlocks from eastern forests. Hemlock dieback can lead to cascading effects on associated ecosystems, including belowground, mycorrhizal fungal communities. Ectomycorrhizal fungi (EM), which are mutualistic with many tree species and provide nutrients to plant hosts, are known to colonize hemlock as well as neighboring tree species at lower levels following HWA infection. This study investigated the effect of hemlock decline from HWA infestation on mycorrhizal communities, as inferred from colonization on northern red oak (Quercus rubra L.) “bait” seedlings grown near “host” hemlock trees. Hemlock health surveys were conducted in healthy (Carl Sandburg Home National Historic Site – CARL) and declining (Warren Wilson College – WWC) stands in western North Carolina, and host trees were paired between stands based on diameter. In each stand, northern red oak seedlings were planted within a meter of host hemlocks in early summer and allowed to grow for 8 w, when they were harvested. Seedling growth and dry biomass were recorded at harvest and roots were sampled for mycorrhizal colonization frequencies. Different mycorrhizal morphotypes were collected from seedling roots for subsequent DNA barcoding analyses to characterize EM taxonomic richness to compare mycorrhizal community assemblages between the two stands. Mycorrhizal colonization frequencies (percentage of the total number of EM-colonized root tips per seedling) and growth in seedling height were significantly greater at CARL than WWC, suggesting healthy hemlock stands are more favorable for oak seedling growth than declining stands. Moreover, a greater proportion of seedlings grown in the healthy stand were colonized by EM, indicating EM assemblages differ between a healthy and a declining hemlock stand. Differences between EM communities corresponded with altered seedling growth allocation, as seedlings in the declining stand had higher root to shoot ratios with reduced stem height, but showed greater investment in root biomass and stem diameter growth. We conclude EM communities differ between a healthy and declining hemlock stands, and changes in EM communities following hemlock dieback may affect the growth of replacement species.