LAUSR

Abstract In eastern North America, the recruitment of native oak species appears to be often low or absent in the presence of Festuca rubra var. rubra, a Eurasian grass. If competitive and allelopathic effects are lowest when plant species share a geographic co-evolutionary history, then oaks may instead grow best in a cover of grasses with which they share a common ancestral habitat. We tested this hypothesis by comparing the impact of F. rubra versus North American native grass (Andropogon gerardii and Elymus virginicus) monocultures on the first two years of growth of North American native oak (Quercus alba, Q. rubra, and Q. macrocarpa) and non-native oak (Q. acutissima, Q. cerris, and Q. robur) seedlings in a field experiment. We predicted the native oaks would grow better among the native grasses than among F. rubra, and that the opposite response would be observed for the non-native oaks. Based on the total biomass data from the second growing season, there was partial support for our prediction, with all three Eurasian oaks growing best among F. rubra, and two out of the three native oaks growing best among the native grass A. gerardii (Q. alba grew poorly among all grass species, possibly due to the suboptimal soil pH for this species). E. virginicus was an exception, in that all oaks grew poorly in its presence, with the exception of Q. robur. Although several of the observed responses could be explained by seasonal variation in the shading effects of the different grass species, the suppressive effect of F. rubra on the growth of native oak seedlings was not explained by shading, which suggests that belowground mechanisms may play an important role in the recruitment failure of North American oaks grown in the presence of this Eurasian grass.