1092 Macro-detritivore identity drives leaf litter diversity effects Veronique C. A. Vos, Jasper van Ruijven, Matty P. Berg, Edwin T. H. M. Peeters and Frank Berendse V. C.A. Vos (veronique.vos@wur.nl), J. van Ruijven and F. Berendse, Centre for Ecosystem Studies, Environmental Sciences group, Wageningen Univ., PO Box 47, NL–6700 AA Wageningen, the Netherlands. – M. P. Berg, Dept of Ecological Science, VU University Amsterdam, De Boelelaan 1085, NL–1081 HV Amsterdam, the Netherlands. – E. T. H. M. Peeters, Centre for Water and Climate, Environmental Sciences, Wageningen Univ., PO Box 8080, NL–6700 DD Wageningen, the Netherlands. he importance of leaf litter diversity for decomposition, an important process in terrestrial ecosystems, is much debated. Previous leaf litter-mixing studies have shown that non-additive leaf litter diversity effects can occur, but it is not clear why they occurred in only half of the studies and which underlying mechanisms can explain these conflicting results. We hypothesized that incorporating the role of macro-detritivores could be important. Although often ignored, macro- detritivores are known to strongly influence decomposition. To better understand the importance of macro-detritivores for leaf litter mixing effects during decomposition, four common leaf litter species were added separately and in two and four species combinations to monocultures of three different macro-detritivores and a control without fauna. Our results clearly show that leaf litter-mixing effects occurred only in the presence of two macro-detritivores (earthworms and woodlice). Application of the additive partitioning method revealed that in the specific combination of woodlice and the presence of a slow-decomposing leaf litter species in the mixture, leaf litter mixing effects were strongly driven by a selection effect. his was caused by food preference of the isopod: the animals avoided the slow decomposing species when given the choice. However, most leaf litter mixing effects were caused by complementarity effects. he potential mechanisms underlying the complementarity effects are discussed. Our results clearly show that that both leaf litter and macro-detritivore identity can affect litter diversity. his may help to explain the conflicting results obtained in previous experiments. he threat of biodiversity loss has inspired a growing amount of literature focusing on the potential effects on the function- ing of ecosystems worldwide (Loreau et al. 2001, Giller et al. 2004, Hooper et al. 2005, Cardinale et al. 2007). While some ecosystem processes, such as primary production in grasslands, have generally shown a positive relationship with biodiversity (see Cardinale et al. 2007 for a meta-analysis), the influence of diversity on other processes is less clear. One important process that needs further attention is decomposi- tion, which is a crucial process driving both nutrient cycling and ecosystem production (Hobbie 1996, Wardle 1997). Earlier work on single leaf litter species showed that the decomposition process is driven by leaf litter quality, the physiochemical environment and the decomposer com- munity (Berg et al. 1993, Gartner and Cardon 2004). However, when leaf litter species are mixed, as would happen naturally, the picture becomes less clear. It has been suggested that leaf litter diversity does not seem to cause predictable changes in decomposition rate (McTiernan et al. 1997, Wardle 1997). Although the majority of studies (  60%) observed a higher decomposition rate when leaf litter was mixed, one third of the work showed either nega- tive or no leaf litter diversity effects (Gartner and Cardon 2004, Hättenschwiler et al. 2005). he question remains what is causing these conflicting results. he following three explanations for leaf litter diversity effects have been suggested (Hättenschwiler et al. 2005): (1) transfer of either nutrients (McTiernan et al. 1997) or specific leaf litter compounds (Hättenschwiler and Vitousek 2000, Hoorens et al. 2003) among leaf litter species, (2) improved microenvironment for microbes and macro-de- tritivores due to a structurally more complex leaf litter layer (Salamanca et al. 1998, Hector et al. 2000), and (3) effects resulting from higher trophic levels such as food preference shown by macro-detritivores (Hättenschwiler and Bretscher 2001, Schädler and Brandl 2005). Importantly, all the mech- anisms above can be directly or indirectly affected by mac- ro-detritivores. Although soil fauna are known to affect the magnitude and direction of diversity effects (Hättenschwiler and Gasser 2005, Schädler and Brandl 2005) and that these effects can vary per species (Heemsbergen et al. 2004), the impact of macro-detritivores has often been ignored in leaf litter diversity studies. his may be one reason why ambigu- ous results in leaf litter mixing studies have been observed, particularly if the decomposition process is top– down regu- lated (Cardinale et al. 2000, Srivastava et al. 2009). Here, we explicitly test the hypotheses that macro- detritivore presence and identity affect decomposition of leaf litter mixtures. Four temperate leaf litter species vary- ing in leaf litter quality were used in monocultures, two Oikos 120: 1092–1098, 2011 doi: 10.1111/j.1600-0706.2010.18650.x © 2011 he Authors. Oikos © 2011 Nordic Society Oikos Subject Editor: Ulrich Brose. Accepted 9 November 2010