1. Herbivory-induced changes in plant resistance and tolerance traits can mediate the interaction between spatially and temporally separated above- and belowground herbivores. However, it is unknown how long the impact of transient herbivory events last or if legacy effects can be detected systemically in the plant and across plant generations. 2. Here, we studied the impact of transient aboveground herbivory by the specialist caterpillar Manduca sexta on the plant traits of wild tobacco Nicotiana attenuata and the consequences for belowground root-knot nematodes Meloidogyne incognita in short-term, long-term, and transgenerational experiments. 3. In the short- and long-term experiments, above- and belowground herbivory had significant, albeit independent, impacts on key plant traits such as plant biomass, nutrient content, and secondary metabolites. In addition, herbivory influenced fitness parameters such as seed yield and quality. In the long-term and transgenerational experiments, transient aboveground herbivory had facilitating effects on nematodes, irrespective of any treatments of the parental plants. We found evidence for transgenerational legacy effects of both above- and belowground herbivory on the resistance of progeny plants against aboveground herbivores. Additionally, belowground herbivory in the parental generation increased the root biomass of progeny plants. 4. Synthesis. Our results suggest that herbivory may have a long-term impact on plant fitness by improving the performance of progeny plants challenged by above- and belowground herbivory.

1. Along with the global decline of species richness goes a loss of ecological traits. Associated biotic homogenization of animal communities and narrowing of trait diversity threaten ecosystem functioning and human well-being. High management intensity is regarded as an important ecological filter, eliminating species that lack suitable adaptations. Belowground arthropods are assumed to be less sensitive to such effects than aboveground arthropods. 2. Here, we compared the impact of management intensity between (grassland vs. forest) and within land-use types (local management intensity) on the trait diversity and composition in below- and aboveground arthropod communities. 3. We used data on 722 arthropod species living above ground (Auchenorrhyncha and Heteroptera), primarily in soil (Chilopoda and Oribatida) or at the interface (Araneae and Carabidae). 4. Our results show that trait diversity of arthropod communities is not primarily reduced by intense local land use, but is rather affected by differences between land-use types. Communities of Auchenorrhyncha and Chilopoda had significantly lower trait diversity in grassland habitats as compared to forests. Carabidae showed the opposite pattern with higher trait diversity in grasslands. Grasslands had a lower proportion of large Auchenorrhyncha and Carabidae individuals, whereas Chilopoda and Heteroptera individuals were larger in grasslands. Body size decreased with land-use intensity across taxa, but only in grasslands. The proportion of individuals with low mobility declined with land-use intensity in Araneae and Auchenorrhyncha, but increased in Chilopoda and grassland Heteroptera. The proportion of carnivorous individuals increased with land-use intensity in Heteroptera in forests and in Oribatida and Carabidae in grasslands. 5. Our results suggest that gradients in management intensity across land-use types will not generally reduce trait diversity in multiple taxa, but will exert strong trait filtering within individual taxa. The observed patterns for trait filtering in individual taxa are not related to major classifications into above- and belowground species. Instead, ecologically different taxa resembled each other in their trait diversity and compositional responses to land-use differences. These previously undescribed patterns offer an opportunity to develop management strategies for the conservation of trait diversity across taxonomic groups in permanent grassland and forest habitats.

Plant–soil feedback (PSF) may affect above-ground higher trophic levels in glasshouse experiments, but evidence from field studies on the relevance of these multitrophic interactions for plant performance is lacking. Therefore, we examined whether PSF effects of several native and invasive plant species occur also in the field and influence plant damage by above-ground herbivores. Root zone soil from an abandoned urban field was used as inocula for the PSF experiment. First, we grew eight urban grassland plant species (five natives and three invasive species) separately in a glasshouse, with soil biota communities conditioned by the respective species itself (‘home soil’) or by a mixture of all other species (‘foreign soil’). After 13 weeks, one cohort of the plants was placed on an urban field in Berlin to assess damage by naturally colonizing herbivores, while another cohort of the plants stayed in the glasshouse. We observed that the extent of the PSF effects differed between the field and glasshouse cohorts of plants. While we found positive PSF responses for five of the eight plant species in the glasshouse, we found no PSF effects in the field. Further, there was no trend that invasive or native species differed in the direction or extent of PSF responses. Concerning the leaf damage by herbivores of the field plants, we found no evidence that the soil history (home vs. foreign soil) affected the effects of above-ground herbivores on the plants. Synthesis. We conclude that PSF effects are more likely to be found under glasshouse conditions. In the field, PSFs seem to play a minor role for the selected urban grassland species. More generally, our study highlights the need to focus on PSFs under natural conditions and in natural communities (including higher trophic levels), which is often overlooked in PSF research.