Mycorrhizal densities decline in association with nonnative plants and contribute to plant invasion

Belowground interactions between herbaceous native species and nonnative species is a poorly understood but emerging area of interest to invasive‐species researchers. Positive feedback dynamics are commonly observed in many invaded systems and have been suspected in California grasslands, where native plants associate strongly with soil mutualists such as arbuscular mycorrhizal fungi. In response to disturbance, invading nonnative plants proliferate, and to the degree these species associate weakly with soil mutualists, we would expect mutualist efficacy to degrade over time. Degraded mutualist efficacy would negatively impact mutualist‐dependent native species or their recruitment following a disturbance. We investigated the feedback dynamics of soil conditioned both with native and nonnative herbaceous communities of southern California grasslands to test this degraded mutualist hypothesis. Using a mesocosm approach, we inoculated each community with live soil originating from a remnant native grassland and varied the plant communities (i.e., native or nonnative) along a plant–species‐richness gradient. After one year, we then used this conditioned soil for reciprocal feedback tests on a native and nonnative indicator species. We show that a native herbaceous forb ( Gnaphalium californicum ) grows best in soil conditioned by a diverse mix of other native species that includes G. californicum but is inhibited by soil conditioned by a diverse mix of nonnative species. We also show that an invasive, nonnative herbaceous forb ( Carduus pycnocephalus ) exhibits strong growth in soil lacking arbuscular mycorrhizal fungi and in soil conditioned by a diverse mix of nonnative species that include C. pycnocephalus , and that it is inhibited by the same soil that best promotes the native, G. californicum . Separate bioassays for mycorrhizal density show a reduction of arbuscular mycorrhizal fungi in the nonnative‐conditioned soil relative to the native‐conditioned soil, which suggests that nonnative species do not promote the growth of mycorrhizal fungi in the same way that native species do. The growth patterns resulting from the vegetative history of these distinct soil communities provide evidence of a biotic feedback mechanism that may account for the maintenance of persistent communities of nonnative (and often invasive) plants ubiquitous throughout California grasslands.