Soil organisms have an important role in aboveground community dynamics and ecosystem functioning in terrestrial ecosystems. However, most studies have considered soil biota as a black box or focussed on specific groups, whereas little is known about entire soil networks. Here we show that during the course of nature restoration on abandoned arable land a compositional shift in soil biota, preceded by tightening of the belowground networks, corresponds with enhanced efficiency of carbon uptake. In mid- and long-term abandoned field soil, carbon uptake by fungi increases without an increase in fungal biomass or shift in bacterial-to-fungal ratio. The implication of our findings is that during nature restoration the efficiency of nutrient cycling and carbon uptake can increase by a shift in fungal composition and/or fungal activity. Therefore, we propose that relationships between soil food web structure and carbon cycling in soils need to be reconsidered.

Earthworm distribution in global soils: Earthworms are key components of soil ecological communities, performing vital functions in decomposition and nutrient cycling through ecosystems. Using data from more than 7000 sites, Phillips et al. developed global maps of the distribution of earthworm diversity, abundance, and biomass (see the Perspective by Fierer). The patterns differ from those typically found in aboveground taxa; there are peaks of diversity and abundance in the mid-latitude regions and peaks of biomass in the tropics. Climate variables strongly influence these patterns, and changes are likely to have cascading effects on other soil organisms and wider ecosystem functions. Science , this issue p. 480 ; see also p. 425

Soil organisms drive major ecosystem functions by mineralising carbon and releasing nutrients during decomposition processes, which supports plant growth, aboveground biodiversity and, ultimately, human nutrition. Soil ecologists often operate with functional groups to infer the effects of individual taxa on ecosystem functions and services. Simultaneous assessment of the functional roles of multiple taxa is possible using food‐web reconstructions, but our knowledge of the feeding habits of many taxa is insufficient and often based on limited evidence. Over the last two decades, molecular, biochemical and isotopic tools have improved our understanding of the feeding habits of various soil organisms, yet this knowledge is still to be synthesised into a common functional framework. Here, we provide a comprehensive review of the feeding habits of consumers in soil, including protists, micro‐, meso‐ and macrofauna (invertebrates), and soil‐associated vertebrates. We have integrated existing functional group classifications with findings gained with novel methods and compiled an overarching classification across taxa focusing on key universal traits such as food resource preferences, body masses, microhabitat specialisation, protection and hunting mechanisms. Our summary highlights various strands of evidence that many functional groups commonly used in soil ecology and food‐web models are feeding on multiple types of food resources. In many cases, omnivory is observed down to the species level of taxonomic resolution, challenging realism of traditional soil food‐web models based on distinct resource‐based energy channels. Novel methods, such as stable isotope, fatty acid and DNA gut content analyses, have revealed previously hidden facets of trophic relationships of soil consumers, such as food assimilation, multichannel feeding across trophic levels, hidden trophic niche differentiation and the importance of alternative food/prey, as well as energy transfers across ecosystem compartments. Wider adoption of such tools and the development of open interoperable platforms that assemble morphological, ecological and trophic data as traits of soil taxa will enable the refinement and expansion of the multifunctional classification of consumers in soil. The compiled multifunctional classification of soil‐associated consumers will serve as a reference for ecologists working with biodiversity changes and biodiversity–ecosystem functioning relationships, making soil food‐web research more accessible and reproducible.

Earthworms and microorganisms are interdependent and their interactions regulate the biogeochemistry of terrestrial soils. Investigating earthworm–microorganism interactions, we tested the hypothesis that differences in burrowing and feeding habits of anecic and endogeic earthworms are reflected by the existence of ecological group-specific gut wall bacterial communities. Bacterial community was detected using automated ribosomal intergenic spacer analysis of 16S and 23S genes and ribotype data was used to assess diversity and community composition. Using soil and earthworm samples collected from adjacent wheat–barley and grass–clover fields, we found that the anecic Lumbricus terrestris and L. friendi, the endogeic Aporrectodea caliginosa and A. longa (classically defined as anecic, but now known to possess endogeic characteristics) contain ecological group-specific gut wall-associated bacterial communities. The abundance of specific gut wall-associated bacteria (identified by sequence analysis of ribotype bands), including Proteobacteria, Firmicutes and an actinobacterium, was ecological group dependent. A microcosm study, conducted using A. caliginosa and L. terrestris and five different feeding regimes, indicated that food resource can cause shifts in gut wall-associated bacterial community, but the magnitude of these shifts did not obscure the delineation between ecological group specificity. Using A. caliginosa and A. longa samples collected in six different arable fields, we deduced that, within an ecological group, habitat was a more important determinant of gut wall-associated bacterial community composition than was host species. Hence, we conclude that the selection of bacteria associated with the gut wall of earthworms is a natural selection process and the strongest determinant of this process is in the order ecological group>habitat>species.