The mutualism between plants and arbuscular mycorrhizal fungi (AMF) is widespread and has persisted for over 400 million years. Although this mutualism depends on fair resource exchange between plants and fungi, inequality exists among partners despite mechanisms that regulate trade. Here, we use (33) P and (14) C isotopes and a split-root system to test for preferential allocation and reciprocal rewards in the plant-AMF symbiosis by presenting a plant with two AMF that differ in cooperativeness. We found that plants received more (33) P from less cooperative AMF in the presence of another AMF species. This increase in (33) P resulted in a reduced (14) C cost per unit of (33) P from less cooperative AMF when alternative options were available. Our results indicate that AMF diversity promotes cooperation between plants and AMF, which may be an important mechanism maintaining the evolutionary persistence of and diversity within the plant-AMF mutualism.

Plant colonization by arbuscular mycorrhizal fungi (AMF) is widespread and can offer considerable benefits in terms of growth, nutrient uptake and plant yield. However, it is still unresolved how different plant species and plant functional groups respond to AMF and to different AMF taxa. Here we established 336 grassland microcosms to determine the response of 14 plant species displaying contrasting functional groups (grasses, legumes and non‐leguminous forbs) for the presence of three different AMF taxa. For each plant species, we calculated the degree to which plant growth depended on AMF colonization (i.e. mycorrhizal dependency [MD]). We also determined the degree to which each plant species relied on specific AMF taxa for optimal growth (i.e. mycorrhizal species sensitivity [MSS]). Additionally, we determined whether MD and MSS correlated to specific plant traits (i.e. specific root length [SRL], specific leaf area [SLA]). The plant growth response to AMF ranged from −84.9% for a non‐mycorrhizal plant (Luzula campestris) to +94.0% for a legume (Trifolium arvensis). The MD was systematically higher in legumes (91.9% ± 2.4%), followed by non‐leguminous forbs (77.1% ± 11.06) and grasses (42.1% ± 15.73%). MSS was less variable (8.9%–37.7%); it was independent of plant functional group and did not correlate with MD. MD was linked to various mycorrhizal plant parameters, including AMF colonization (R2 = +0.80) and total dry biomass (R2 = +0.32). Moreover, among mycorrhizal plants (n = 12), MD negatively correlated with SRL (R2 = −0.24) and positively with SLA (R2 = +0.24). Synthesis. This study shows that plants relying on AMF for biomass production also show higher root colonization, lower SRL, higher SLA and that different plant traits are interlinked with the way how plants respond to AMF. Overall, this study further demonstrates that different plant functional groups vary in their response to AMF. Read the free Plain Language Summary for this article on the Journal blog.