Abstract Aims Plant–soil feedbacks (PSFs) are key drivers of forest composition and diversity, yet their direction and magnitude may depend on the mycorrhizal identity of interacting species, environmental conditions, and experimental context. Methods We conducted complementary greenhouse and field experiments using Acer rubrum L., Acer saccharum Marsh., Prunus serotina Ehrh. (all arbuscular mycorrhizal AM species), and Quercus alba L., and Quercus rubra L. (all ectomycorrhizal EM species) to test how biomass-based PSFs vary with mycorrhizal matching between seedlings and adult trees, light availability, and soil microbial communities. Seedlings were grown in soil conditioned by conspecifics, conmycorrhizal heterospecifics, or heteromycorrhizal heterospecifics under controlled and natural light regimes. Results Consistent with expectations, AM species consistently exhibited negative PSFs under low light, and EM species tended to show more positive PSFs, but this pattern was contingent on light and greenhouse versus field setting. For AM species, negative PSFs occurred primarily under low light and were neutralized or reversed under higher light. EM species showed generally more positive PSFs across light levels and settings, although species-specific differences emerged. PSFs were driven largely by conspecific soil conditioning, with limited influence from the mycorrhizal identity of heterospecific neighbors. Results from greenhouse versus field settings diverged, with field PSFs sometimes attenuated or reversed, particularly for Q. alba. Conclusions These findings highlight that biomass-based PSFs are not fixed species traits but context-dependent outcomes influenced by mycorrhizal type, light availability, and environmental setting. Incorporating these factors is essential for predicting how PSFs influence seedling recruitment, forest dynamics, and biodiversity.
McCarthy-Neumann et al. (2026) studied this question.