Endemic species of ectomycorrhizal fungi (EMF) can be found throughout many forest biomes, but it is unclear whether their localized distribution is dictated by deterministic processes or geographical barriers to dispersal. We investigated the adaptive characteristics and prevalence of endemic versus cosmopolitan EMF species in perhumid temperate rainforests of southwestern Vancouver Island (Canada), characterized by moist, acidic soils with high nitrogen (N) supply alongside low phosphorus (P) and cation availability. Endemic EMF species, representing almost half of the community, had significantly higher sporocarp N (24% increase), potassium (+16%), and magnesium (+17%) concentrations than cosmopolitan species. Sporocarp P concentrations were low overall, reflecting limited soil P availability, and did not differ by fungal range. However, sporocarp N% and P% were well correlated, and species with higher N concentrations showed an increasing N:P ratio, supporting evidence for the N allocation required to produce organic P-acquiring enzymes. Endemics were also more likely to occur on Tsuga heterophylla (a disjunct host genus) than Picea sitchensis (a circumpolar genus), but pairwise comparisons indicated no differences in abundance by fungal range for either host. Endemics represented a diverse group, with moderate dispersion across the phylogeny. The Inocybaceae and Thelephoraceae families had high proportions of endemic taxa, while Cortinariaceae was largely cosmopolitan, highlighting some niche conservatism in certain lineages but not as an overall pattern. We conclude that superior adaptive traits in relation to perhumid soils were skewed towards the endemic community, underscoring the important contribution of these localized fungi to rainforest nutrition and productivity. 

In the fall of 2020, over a six-week period (October 5-November 15) we visited each plot three times to search for fleshy, terricolous sporocarps. We selected sporocarps that were fresh and free from insect damage and strove to obtain composite samples for each fungal species, typically 3–6 individual sporocarps, from over the entire plot area. Following collection, sporocarps were cleaned of debris and dried in a 60°C convection oven for 24 h (larger sporocarps were sliced in half to facilitate drying). Species identity was confirmed by molecular DNA of the fungal ITS region using a small piece of dried sporocarp, and the remainder of the sample (cap and stipe) was ground to < 2 mm for nutritional analysis.

In June 2021, soil cores approximately 12 cm in diameter and 15 cm in depth were removed with a soil knife at 15 random locations in each of the ten plots. The soil core included upper mineral soil and organic horizons. Samples were returned to the laboratory and kept at 4°C during processing. Each core was submerged under water in a large pan and random portions of roots removed for examination under a dissecting microscope. Fresh, vigorous colonies of EMF root tips were detached using tweezers. Composited root tips from three soil cores were combined into one vial, for a total of 5 subsamples per plot (n = 50 for the study). This process was conducted by two personnel for 30 minutes per soil core, and resulted in 7–10 mL of fresh root tips per vial. The samples were then freeze-dried and ground with a ball mill before DNA extraction.