Although trophic cascades (in which feeding by top predators induces a chain of effects along the food chain that affects lower trophic levels) have been well documented in several aquatic systems15xConsumer control of lake productivity. Carpenter, S.R. and Kitchell, J.F. BioScience. 1988; 38: 764–769CrossrefSee all References15, the evidence for cascades in terrestrial food webs remains much more equivocal. If trophic cascades are important in soil food webs, the question emerges as to what effect this has on the ecosystem-level functions performed by the organisms in those lower trophic levels affected by the cascade. The study of Laakso and Setala6xSensitivity of primary production to changes in the architecture of belowground food webs. Laakso, J. and Setala, H. Oikos. 1999; 87: 57–64CrossrefSee all References6provides clear evidence that lower trophic levels in soil food webs are much more important than higher ones in regulating a key ecosystem function (plant productivity), which would suggest that trophic cascades do not occur or, if they do, that the effects are not detectable at the ecosystem level of resolution. There have been very few studies that have looked at the ecosystem-level consequences of manipulation of the higher trophic levels of soil food webs, although two earlier studies provide relevant data. In the first of these, Santos and colleagues16xThe role of mites and nematodes in early stages of buried litter decomposition in a desert. Santos, P.F., Phillips, J., and Whitford, W.G. Ecology. 1981; 62: 664–669CrossrefSee all References16 found that reducing populations of predatory tyleid mites produced the following chain of events: •Reduce tyleid mites → increase bacterial-feeding nematodes → reduce bacteria → reduce decomposition rateIn the second study, Kajak and colleagues17xExperimental studies on the effects of epigeic predators on matter decomposition processes in managed peat grasslands. Kajak, A. et al. Pol. Ecol. Stud. 1993; 17: 289–310See all References17 found that exclusion of epigeic predatory arthropods had the following consequences: •Reduce predatory arthropods → increase saprophagous mesofauna → increase microflora → increase decomposition and N mineralization ratesBased on the limited data available, it would appear that the nature of ecosystem-level effects of top predators resulting from trophic cascades are probably case-specific. In the Laakso and Setala6xSensitivity of primary production to changes in the architecture of belowground food webs. Laakso, J. and Setala, H. Oikos. 1999; 87: 57–64CrossrefSee all References6 study at least, trophic cascades were found to be weak or nonexistent (and to have no detectable effects above-ground) despite top predators significantly reducing their prey. Although it is likely that top predators can indirectly affect the rate at which microbes are consumed in the microcosms, the microflora in these systems can probably compensate for biomass consumption through alteration of its rate of turnover18xNo evidence for trophic cascades in an experimental microbial-based soil food web. Mikola, J. and Setala, H. Ecology. 1998; 79: 153–164CrossrefSee all References18. This should, in turn, help buffer microbially mediated processes against effects of top predators. Whether this pattern holds in situations with very different top predators, plant species and substrates remains mostly unknown, but the study of Laakso and Setala6xSensitivity of primary production to changes in the architecture of belowground food webs. Laakso, J. and Setala, H. Oikos. 1999; 87: 57–64CrossrefSee all References6 provides an experimental framework within which such effects could be tested.