Plants and mycorrhizal fungi form close mutualistic relationships that affect the structure and function of ecosystems. Recent studies show how certain plant communities can form associations with the same mycorrhizal fungus leading to preferential transfer of carbon and limiting nutrients, such as nitrogen, between different species through common mycorrhizal networks. This phenomenon has been documented in a wide range of ecosystems but its mechanisms remain poorly understood. In an experimental prairie-pasture plant system, replicated at three sites across a 520 km latitudinal gradient in the Pacific Northwest, USA, we identified general processes that reconcile competing hypotheses related to resource transfer in common mycorrhizal networks. We used stable isotope tracers, paired with analyses of fungal DNA and plant and fungal community structure and traits, to quantify above- and belowground allocation and transfer of carbon and nitrogen between 18 different plant species. We applied isotopically enriched carbon (13C) and nitrogen (15N) as a pulse of carbon dioxide (CO2) and ammonia (NH3) to the leaves of target 'donor' species common across experimental sites. At each site, we performed >1360 measurements of foliar and root isotopic enrichment, which revealed morphological traits and tissue stoichiometry as the most important predictors of interspecific resource transfer. Assimilation of isotopic tracers was similar between labeled 'donor' plants with no significant difference between sites for labeled 'donor' plants with no significant difference in transfer between species among experimental treatments, including rainfall exclusion or species diversity. However, allocation of carbon and nitrogen to roots and transfer to 'receiver' species varied significantly between functional groups (annual/perennial or grass/forb) due to differences in tissue stoichiometry. Our findings points to a simple mechanistic answer for long-standing questions regarding mutualism and transfer of resources between plants via mycorrhizal networks, an explanation that can lead to general predictions of preferential flow of limiting resources in other ecosystems.