Abstract
Nitrogen (N) additions have decreased species richness ( S ) in hardwood forest herbaceous layers, yet the functional mechanisms for these decreases have not been explicitly evaluated. We tested two hypothesized mechanisms, random species loss ( RSL ) and non‐random species loss ( NRSL ), in the hardwood forest herbaceous layer of a long‐term, plot‐scale, fertilization experiment in the central Appalachian Mountains, USA . Using a random thinning algorithm, we simulated changes in species densities under RSL and compared the simulated densities to the observed densities among N‐fertilized (+N), N‐fertilized and limed (+N+L), and reference ( REF ) plots in regenerating forest stands. We found a lower S in the +N treatment across all survey years and determined that the reduction in S was a function of NRSL . Furthermore, non‐random effects were observed in certain species, as they occurred at densities that were either higher or lower than expected due to RSL . Differential advantages were also observed among species between +N and +N+L treatments, suggesting that species responded to either the fertilization or acidification effects of N, though no consistent pattern emerged. Species nitrophily status was not a useful trait for predicting specific species losses, but was a significant factor when averaged across all treatments and sampling years. Our results provide strong evidence that declines in S in the forest herbaceous layer under N fertilization are due largely to NRSL and not simply a function of species rarity.