1. Increasing evidence suggest that plant-soil interactions play an essential role in plant community assembly processes. Empirical investigations show that plant species abundance in the field is often related to plant-soil biota interactions; however, the direction of these relations have yielded inconsistent results. 2. We combined unique 31-year long field data on species abundances from a species-rich mountain meadow with single time point plant-soil feedback greenhouse experiments of 24 co-occurring plant species. We tested whether these relations were dynamic in time, whether coupled increases and decreases in abundance between years were related to plant-soil feedback and whether these changes were underlain by years in which manuring was applied. 3. The prevailingly negative relationship between plant-soil feedback and plant relative abundance in the field was significantly time-dependent, which may reconcile the contrasting results in literature. Furthermore, significantly coupled oscillations appeared between species relative abundance changes and plant-soil feedback, which were likely moderated by years in which manuring was applied. Our results are consistent with the notion that the more abundant species are stabilised by negative plant-soil feedback, and the less abundant species co-vary with the fluctuations of these more competitive species. 4. Synthesis. Our results project plant-soil feedback as an important regulatory mechanism in plant communities, operating in conjunction with species' competitive ability and soil nutrient availability. We suggest that negative feedback is particularly prominent in more abundant plant species that profit from more readily available soil nutrients than less abundant species with positive feedback. Negative plant-soil feedback may thus prevent more abundant plant species from out-competing less abundant plant species, facilitating stable species coexistence .

Background and aims The concept of plant-soil feedback is increasingly used to explain plant community assembly processes. Soil nutrient availability can be expected to play a critical role on these processes. However, little is known about the effects of nutrient availability on feedback direction and strength. Methods A plant-soil feedback experiment was performed with the grasses Anthoxanthum odoratum and Festuca rubra, and the forbs Leontodon hispidus and Plantago lanceolata, on soil with either low or high nutrient availability. Additionally, we tested if plantsoil feedback of the two forbs under these conditions changed by inoculation of the soil with spores of an arbuscular mycorrhizal fungus. Results Increased nutrient availability neutralised plantsoil feedback based on shoot biomass independent of its negative or positive direction, whereas the effects on root biomass were either not altered or turned negative. Mycorrhizal fungi spore addition decreased negative feedback and increased positive feedback. Conclusions Our results suggest that negative plant-soil feedback on low nutrient soil can be overcome with nutrient addition, and that positive soil biota associations on low nutrient soil may become superfluous with nutrient increase. We hypothesize that species-specific, microbial mediated plant community assembly processes occur in low rather than high nutrient environments.

HighlightPatches rich in nitrogen are rapidly colonized by selective root growth in maize, which was quantified at high time resolution with state-of-the-art non-invasive imaging techniques in a paper-based growth system.

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

Take-down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Interactions between plants and soil biota are increasingly shown to play critical roles in plant species coexistence processes. Plant species coexistence is thought to be promoted via biotic legacies that plant species leave behind in the soil after a plant disappears. These soil legacies are hypothesised to supress colonisation success when the preceding plant is of the same species, that is, when a plant species encounters its own, species‐specific soil antagonists. However, colonisation of vacant spots in plant communities is in the first place determined by the ability of plants to reach such vacant locations. We currently lack an understanding of the explicit role of soil legacy effects and their relative contribution to colonisation processes in plant communities consisting of plant species inherently differing in colonisation ability. In experimental, outdoor plant communities consisting of eight grassland species, we tested the effect of five differently conditioned soil patches on plant species colonisation success over three consecutive growing seasons. We found that colonisation success was largely determined by the species' reproductive strategy, lateral spread and growth rate, and not by the plant species that conditioned the soil patch. Fast spreading, clonal plant species reached the soil patches first and initially attained the highest biomass inside the patch. One year later, slower spreading plant species colonised the patch via seedlings. Species with intrinsically high growth rates attained the highest biomass, decreasing biomass of the initial colonisers. While subtle differences between conditioned soil patches did occur, these were not strong enough to overcome the inherent differences in colonisation ability between the various plant species. Synthesis. Our results reject the hypothesis that colonisation of vacant soil patches in plant communities is strongly affected by the legacy that is left behind by the preceding plant species. Instead, plant species life‐history strategy plays a prominent role, driving sequential plant species replacements. Based on our results and recent accounts in literature we present a conceptual model for local cyclic dynamics in grassland communities, where soil legacy plays a role in affecting the performance of established plant species rather than colonisation of vacant patches.

Soil microbial networks play a crucial role in plant community stability. However, we lack knowledge on the network topologies associated with stability and the pathways shaping these networks. In a 13-year mesocosm experiment, we determined links between plant community stability and soil microbial networks. We found that plant communities on soil abandoned from agricultural practices 60 years prior to the experiment promoted destabilising properties and were associated with coupled prokaryote and fungal soil networks. This coupling was mediated by strong interactions of plants and microbiota with soil resource cycling. Conversely, plant communities on natural grassland soil exhibited a high stability, which was associated with decoupled prokaryote and fungal soil networks. This decoupling was mediated by a large variety of past plant community pathways shaping especially fungal networks. We conclude that plant community stability is associated with a decoupling of prokaryote and fungal soil networks and mediated by plant-soil interactions.

Soil microbial networks play a crucial role in plant community stability. However, we lack knowledge on the network topologies associated with stability and the pathways that shape these networks. In a 13-year mesocosm experiment, we determined how natural grassland soil and soil abandoned from agricultural practices 60 years before the start of the experiment affected soil microbial network topologies. Abandoned arable soil promoted destabilising properties both above- and belowground. Aboveground, instability was associated with invading plant species reaching dominance. Belowground, instability was associated with soil microbial networks coupled in prokaryote and fungal responses, which were both shaped by a few, dominating plant community parameters. Conversely, in stable, natural grassland communities, soil prokaryote and fungal responses were decoupled. This decoupling was associated with different sets of plant community parameters shaping prokaryote and fungal niches. We conclude that plant community stability is associated with soil microbial networks with a high niche differentiation.