Which ecological processes assemble diverse floral fungal communities across habitat and host boundaries remains an open question. Three ecological hypotheses that may explain fungal diversity are: (H1) species sorting and environmental filtering when fungal relative abundances differ by plant species, (H2) plant and pollinator diversity begets fungal diversity, and (H3) cross-habitat fungal spillover. Using amplicon sequencing of the ITS1 region, we tested these hypotheses in floral fungi communities for three generalist wildflower species in patchy habitat mosaics of serpentine and non-serpentine grasslands. Our findings suggest some evidence for our first and third hypotheses. In accordance with our first hypothesis, yeasts were more abundant on clover, while filamentous fungi were more abundant on goldfields and vetch and overall community composition varied between host species. In contrast to our second hypothesis, plant and pollinator diversity did not correlate with fungal diversity except for with the aster goldfields. In accordance with our third hypothesis, wildflower species in different habitats shared abundant fungal taxa, suggesting evidence of fungal spillover across habitats. Species sorting and fungal spillover may interact to assemble floral fungal communities, challenging assumptions that plant and pollinator diversity beget fungal diversity.

Plant-pollinator mutualisms are essential to maintain biodiversity, but anthropogenic changes are causing their rapid decline. While diverse pollinators are supported by spatially heterogenous landscapes, understanding how pollinators spill over across habitats to maintain mutualisms is an emerging research area. Theoretical models of mutualisms often ignore such cross-habitat spillover. We ask: does cross-habitat spillover maintain mutualisms and populations of mutualists? To address this gap, we demonstrate how foundational concepts from spatial ecology—metacommunities and resource subsidies—can strengthen mutualistic theory. This Perspective paper advances a novel, conceptual framework for elucidating how spillover across ecological boundaries affects mutualistic diversity, structure, function, and dynamics. Using a classic Lotka-Volterra model, we demonstrate how spatial concepts can integrate into mutualistic theory. While we focus on plant-pollinator mutualisms due their conservation relevance, we suggest questions for future work that could be extended to other types of mutualisms. Themes for future research include examining how additional community interactions and anthropogenic changes drive how spillover affects mutualisms. Ultimately, we recommend deeper integration of niche-based theory with metacommunity and resource-subsidy dynamics in the study of mutualisms.