Hostname: page-component-745bb68f8f-f46jp Total loading time: 0 Render date: 2025-02-11T10:33:42.532Z Has data issue: false hasContentIssue false
  • English
  • Français

Trophic guilds of generalist feeders in soil animal communities as indicated by stable isotope analysis (15N/14N)

Published online by Cambridge University Press:  29 January 2010

K. Oelbermann  and
S. Scheu
K. Oelbermann
Affiliation:
Darmstadt University of Technology, Institute of Zoology, Schnittspahnstr. 3, 64287Darmstadt, Germany
S. Scheu*
Affiliation:
Georg-August-University Göttingen, J.F. Blumenbach Institute of Zoology and Anthropology, Berliner Str. 28, 37073Göttingen, Germany
*
*Author for correspondence Fax: 0049-551-395448 E-mail: sscheu@gwdg.de
Rights & Permissions [Opens in a new window]

Abstract

We investigated if the commonly used aggregation of organisms into trophic guilds, such as detritivores and predators, in fact represent distinct trophic levels. Soil arthropods of a forest-meadow transect were ascribed a priori to trophic guilds (herbivores, detritivores, predators and necrovores), which are often used as an equivalent to trophic levels. We analysed natural variations in 15N/14N ratios of the animals in order to investigate the trophic similarity of organisms within (a priori defined) trophic guilds. Using trophic guilds as an equivalent to trophic level, the assumed stepwise enrichment of 15N by 3.4‰ per trophic level did not apply to detritivores; they were only enriched in 15N by on average 1.5‰ compared to litter materials. Predators on average were enriched in 15N by 3.5‰ compared to detritivores. Within detritvores and predators δ15N signatures varied markedly, indicating that these trophic guilds are dominated by generalist feeders which form a gradient of organisms feeding on different resources. The results indicate that commonly used trophic guilds, in particular detritivores and predators, do not represent trophic levels but consist of subguilds, i.e. subsets of organisms differing in resource utilization. In particular, in soil and litter food webs where trophic level omnivory is common, the use of distinct trophic levels may be inappropriate. Guilds of species delineated by natural variations of stable isotope ratios are assumed to more adequately represent the structure of litter and soil food webs allowing a more detailed understanding of their functioning.

Keywords

soil animal communitytrophic level15Ntrophic level enrichmentdetritivorespredators
Type
Research Paper
Information
Bulletin of Entomological Research , Volume 100 , Issue 5 , October 2010 , pp. 511 - 520
Copyright
Copyright © Cambridge University Press 2010

Introduction

Trophic links in food web studies often are hierarchically structured according to distinct trophic levels (Hunt et al., Reference Hunt, Coleman, Ingham, Inham, Elliott, Moore, Rose, Reid and Morley1987; Schaefer, Reference Schaefer1990; De Ruiter et al., Reference De Ruiter, Neutel, Moore, Polis and Winemiller1996; Zheng et al., Reference Zheng, Bengtsson and Agren1997; Bengtsson et al., Reference Bengtsson, Lundkvist, Saetre, Sohlenius and Solbreck1998). Trophic levels usually are assumed to consist of animal guilds feeding on similar resources with the distance of adjacent levels being equivalent to that between a consumer and its resource. In terrestrial animal communities, consumers feeding on dead organic matter are usually aggregated to detritivores and those feeding on living plant material to herbivores. Consumers feeding on living animal tissue are aggregated to predators and those feeding on dead animals to necrovores. These trophic guilds often are used as equivalents to trophic levels.

Ascribing species to trophic levels is particularly difficult in soil animal communities in which generalist opportunistic feeders and trophic level omnivores predominate (Scheu, Reference Scheu2002; Scheu & Setälä, Reference Scheu, Setälä, Tscharntke and Hawkins2002). In terrestrial ecosystems, a substantial part of primary production enters the detritus food web (Polis, Reference Polis1991; Hairston & Hairston, Reference Hairston and Hairston1993; Coleman & Crossley, Reference Coleman and Crossley1996). Primary decomposers, such as fungi and bacteria, dissipate most of the detritus bound energy. Abundant decomposing arthropods, such as Collembola and Diptera, are considered to predominantly feed on fungi gaining energy out of detritus associated microorganisms (Schaefer, Reference Schaefer, Röhrig and Ulrich1991). The diet of detritivores includes detrital materials originating from all trophic levels of the food web, i.e. plant and animal tissues, bacteria, algae and fungi. Detritivores differently use this heterogeneous pool of resources; and, as a consequence, the trophic guild of detritivores consists of subguilds of species forming a gradient of increasing trophic position spanning over more than one trophic level. The same likely is true for epigeic predators hunting on the soil surface, such as lycosid spiders, carabid and staphylinid beetles, for which detritivorous arthropods are an important food resource (Swift et al., Reference Swift, Heal and Anderson1979; Eisenbeis & Wichard, Reference Eisenbeis and Wichard1985; Nyffeler et al., Reference Nyffeler, Sterlling and Dean1994; Halaj & Wise, Reference Halaj and Wise2002). The trophic guild of predators in detritus-based food webs may consist of trophic subguilds differing in the relative contribution of detritivores, herbivores and predators to their diet. Further, feeding on different subguilds of detritivores or different amounts of intraguild prey or plant material suggests the existence of trophic subguilds within predators.

Besides trophic niche separation among generalist feeders, switching between resources reinforces the difficulty to assign generalist feeders to trophic levels. Also, generalist feeding results in a wide spectrum of potential interactions, which often are not adequately represented in food web models (Polis et al., Reference Polis, Myers and Holt1989). There is growing evidence that the concept of distinct trophic levels may not apply to soil animal communities in which generalist feeders predominate (Ponsard & Arditi, Reference Ponsard and Arditi2000; Scheu & Falca, Reference Scheu and Falca2000; Scheu, Reference Scheu2002). For this reason, it has been suggested to abandon the trophic level concept (Polis & Strong, Reference Polis and Strong1996). In particular, detritivores and predators in soil likely consist of different trophic guilds spanning over more than one trophic level. A similar problem exists in taking taxonomic groups, such as spiders, as an equivalent of trophic species.

The analysis of natural variations in stable isotope ratios of nitrogen in animal tissue is increasingly used to analyse the structure of soil food webs (Ponsard & Arditi, Reference Ponsard and Arditi2000; Scheu & Falca, Reference Scheu and Falca2000; McNabb et al., Reference McNabb, Halaj and Wise2001; Schmidt et al., Reference Schmidt, Curry, Dyckmans, Rota and Scrimgeour2004; Albers et al., Reference Albers, Schaefer and Scheu2006). The method is based on the assumption of constant 15N enrichment per trophic level by 3.4‰, which has been established by analysing aquatic and above-ground terrestrial food chains (Minagawa & Wada, Reference Minagawa and Wada1984; Post, Reference Post2002). At present, there is only little experimental evidence whether the factor of 3.4 δunits is also true for soil invertebrates. As stressed by Gannes et al. (Reference Gannes, O'Brien and Martinez del Rio1997) and Vanderklift & Ponsard (Reference Vanderklift and Ponsard2003) the fractionation in 15N per trophic level may not be constant. Enrichment in 15N appears to be higher in predators feeding on protein rich diets than in herbivores living on low nitrogen food (Vanderklift & Ponsard, Reference Vanderklift and Ponsard2003). Low enrichment may also apply to detritivores, but there is little evidence whether this is true (Ponsard & Arditi, Reference Ponsard and Arditi2000; Scheu & Falca, Reference Scheu and Falca2000; Vanderklift & Ponsard, Reference Vanderklift and Ponsard2003).

The present study uses the stable isotope methodology to test if aggregation of organisms into trophic guilds, such as detritivores and predators, in fact represent groups of trophic similarity. We tested if fractionation of 15N in detritivores is different from the postulated 3.4‰ (Minagawa & Wada, Reference Minagawa and Wada1984), assuming that the enrichment in 15N in consumers depends on nitrogen content of their resources. We hypothesised that food webs with high degrees of omnivory and opportunistic feeding consist of a gradient of organisms consuming different subsets of resources. Therefore, we expected both detritivores and predators to consist of trophic subguilds forming a gradient of increasing trophic position rather than distinct trophic levels.

Materials and methods

The arthropod community of a forest-meadow transect in the Kranichsteiner Wald near Darmstadt (Hessen, Germany) was studied. The transect studied included a forest site, a meadow site and the transition area (distance between sites 20–40 m). Ten open pitfall traps were placed in each of the sites at a distance of 5 m from the next. Pitfall traps constituted of glass jars of a height of 12 cm and a diameter of 5.5 cm. The traps were evenly connected to the soil surface by a plastic ring and filled with ca. 50 ml of a 1:1 glycerol-water solution.

From May to October 2001 and from March to April 2002, the traps were operated for a period of two weeks at monthly intervals. Animals caught were transferred into 70% alcohol and stored until determination and counting. The 15N content of dominant arthropod species was analysed. Dominance estimates were based on numbers of arthropods per catch.

Selection of arthropods for stable isotope analysis was based on three criteria: (i) we selected arthropods out of each of the trophic groups (detritivores, herbivores, predators, necrovores); (ii) we selected species which we expected to be linked by predator – prey interactions (e.g. collembolans and spiders); and (iii) we restricted the analysis to species which allowed replicated analysis of stable isotope ratios.

Preparation of samples

Litter materials and animals were dried at 60°C and then ground with a mortar and pestle. Between 0.24 and 1.57 mg of animal tissue and about 4 mg of litter material were placed in 8×5 mm tin capsules. In large species (ca. >0.5 mg body weight), one individual was used per sample; whereas, in small species (<0.5 mg body weight), several individuals had to be combined. For each species, three replicates were analysed, except for some species which were replicated only twice. Isotope ratios were determined by a coupled system of an elemental analyser (NA 1500, Carlo Erba, Milan) and a mass spectrometer (MAT 251, Finnigan: Reineking et al., Reference Reineking, Langel and Schikowski1993). The stable isotope composition of 15N (δ15N) was calculated as δ15N=((Rsample/Rstandard)−1)×1000, where Rsample is the 15N/14N ratio of the sample and Rstandard is the respective ratio of the standard (Peterson & Fry, Reference Peterson and Fry1987). Atmospheric nitrogen served as primary standard and acetanilid (C8H9NO, Merck, Darmstadt) for internal calibration.

Calculations and statistical analysis

Litter material was assumed to represent the base of the food web. In order to test the usually assumed stepwise enrichment in 15N per trophic level by 3.4‰, animals were ascribed a priori to trophic guilds (detritivores, herbivores, predators, necrovores) commonly used as an equivalent to trophic levels. The assignment was based on published data on the diet of the taxa (Locket & Millidge, Reference Locket and Millidge1951; Freude et al., Reference Freude, Harde and Lohse1976; Fjellberg, Reference Fjellberg1980; Zahradnik, Reference Zahradnik1985; Jones, Reference Jones1990; Heimer & Nentwig, Reference Heimer and Nentwig1991; Schaefer, Reference Schaefer1992; Chinery, Reference Chinery1993; Roberts, Reference Roberts1995; Wachmann et al., Reference Wachmann, Platen and Barndt1995; Sauer, Reference Sauer1996, Reference Sauer1998; Dücker et al., Reference Dücker, Schmüser, Heubel, Borcherding, Heubel, Müller-Reich, Pahnke, Gienapp, Nötzold and Nötzold1997; Harde & Severa, Reference Harde and Severa1998; Witt, Reference Witt1998). We tested if the δ15N signatures differ between the a priori defined trophic guilds by using analysis of variance (ANOVA). Tukey's honestly significant difference test (equal sample size) or Scheffè test (unequal sample size) were used for comparison of means. For estimating the width of trophic guilds, we used the difference between maximum and minimum δ15N signature of the species.

For aggregating species into trophically homogeneous subgroups, we calculated statistically homogeneous subsets of δ15N signatures. For this, we used analysis of variance (ANOVA) with δ15N signatures as dependent and species as independent factor. Homogeneous subgroups were calculated using Scheffé test.

Results

Selection of organisms for analysis of stable isotope ratios

Among herbivores, only Cicadellidae (mainly juvenile stages) and Delphacidae were caught in numbers allowing replicated analysis of stable isotopes.

Detritivores analyzed covered the full range of detritivore taxa with a focus on Collembola since they were most abundant and presumably of particular importance as prey for predators such as spiders. Among Collembola, we focussed on Entomobryidae, the most abundant group (72% of total). Large decomposers were represented by Diplopoda and Isopoda. Further, we analysed Diptera species (Drosophilidae, Sciaridae, Cecidomyiidae, Phoridae) since, as larvae, many of them live in soil but, as adults, may form a substantial part of the diet of aboveground predators. Other detritivores were included due to their omnipresence.

Among predators, we focussed on generalist predators, which are assumed to consume herbivorous and detritivorous prey, i.e. predaceous beetles and spiders. In spiders, we analysed each species caught, in numbers allowing replicated analysis of stable isotopes. More than 80% of the spiders were free-hunting taxa, in particular Lycosidae. In this group, we differentiated adult and juvenile stages. In Coleoptera, we focussed on Staphylinidae and Carabidae. Scydmaenidae and Dytiscidae were included due to their high numbers. Other generalist predators, which were often caught and, therefore, included in the analysis, were Erythraeidae, Phalangiidae, Nemastomatidae, Neobisiidae, Lithobiidae and Asilidae.

At the study site, three necrovores belonging to Sarcophagidae (Diptera), Silphidae and Catopidae (Coleoptera) were caught in high numbers and, therefore, included in the analysis.

A priori defined trophic guilds

Epigeic soil arthropods of the studied forest-meadow transect were dominated by detritivores (80.5% of total sample) and predators (16.0%). Herbivores and necrovores represented 1.4% and 0.7% of total arthropods, respectively. We disregarded the remaining 1.4% of total arthropods (including, e.g. Geotrupidae and Gryllotalpidae), as these unlikely contribute substantially to predator nutrition. Signatures of δ15N of arthropods spanned over 12.6 δunits, ranging from –5.4‰ (Orchesella flavescens, Entomobryidae, Collembola) to 7.2‰ (Sarcophagidae, Diptera).

Mean δ15N signatures of basal resources (litter) and trophic guilds (detritivores, herbivores, predators, necrovores) differed significantly (ANOVA: F 4,244=50.07; P<0.0001). Litter materials, herbivores and detritivores had lowest δ15N signatures, with the mean δ15N signatures of litter materials being similar to those of herbivores and detritivores. Detritivores as trophic guild were enriched in 15N compared to litter by on average 1.5‰. Predators were significantly enriched in 15N compared to litter, herbivores and detritivores, compared to herbivores and detritivores by 5.0‰ and 3.3‰, respectively. Necrovores had highest 15N signatures, being significantly enriched in 15N compared to predators, herbivores and detritivores (fig. 1).

Fig. 1. Mean δ15N signatures (±SD) of litter (L), herbivores (H), detritivores (D), predators (P) and necrovores (N) in a forest-meadow transect. n, the number of species representing this group. Significant differences are represented by different letters (Tukey's HSD test, P<0.05). For aggregation of taxa, see Appendix.

Separation of a priori defined trophic guilds into subguilds

Signatures of δ15N within a priori defined trophic guilds of detritivores and predators varied strongly, indicating that they spread over more than one trophic level. Within detritivores, signatures of δ15N varied by 8.3‰ and within predators by 6.1‰. Species of both detritivores and predators formed gradients differing in δ15N signatures (fig. 1).

Signatures of δ15N of species differed significantly (ANOVA: F 71,177=20.37; P<0.0001). According to the Scheffé test, detritivores were separated into seven (D1–D7), herbivores into three (H1–H3), predators into four (P1–P4), and necrovores into three (N1–N3) subguilds. For species composition of subguilds, see Appendix.

Trophic subguilds of detritivores and predators

Detritivore subguilds D1, D2 and D3 were depleted in 15N compared to litter 15N (Appendix). Signatures of δ15N of detritivores D4 were similar to those of the mean litter 15N and detritivores D5, D6 and D7 were enriched in 15N compared to mean litter 15N.

Predators P1 were enriched in 15N compared to detritivores D1–D4 and to herbivores H1–H3 but depleted compared to detritivores D6–D7. Signatures of δ15N of predators P1 resembled those of detritivores D5 (Appendix). Predators P2 were enriched in 15N compared to detritivores D1–D6 and to herbivores H1–H3, but depleted compared to detritivores D7. Predators P3 and P4 were enriched in 15N compared to each of the detritivore and herbivore subguilds and also to predators P1 and P2.

Both detritivore and predator subguilds included taxonomically similar species. Collembola were distributed among detritivore subguilds D1, D2 and D5 (Appendix); Diplopoda were included in D2, D3, D4 and D5. Within Glomeridae different developmental stages were distributed over different detritivore subguilds. Isopoda were spread over D5 and D6. As in detritivores, different predator subguilds also included taxonomical similar species. Carabid beetles were distributed over P1, P2 and P3 and spiders over P2, P3 and P4. Within spiders, different taxa of Lycosidae and even different developmental stages of two lycosid species (T. terricola and P. pullata) were distributed among different predator subguilds.

Discussion

Trophic guilds

The epigeic animal community of the forest-meadow transect investigated in this study was dominated by detritivorous and predatory arthropods; compared to these groups, herbivores were rare. This suggests that the predominant pathway of the flux of energy is from litter material to detritivores to predators. Based on the dominance of decomposers, litter material was assumed to represent the base of the food web (see also Oelbermann et al., Reference Oelbermann, Langel and Scheu2008). Treated as trophic guild, detritivores in the present study were only enriched in δ15N by, on average, 1.5‰ compared to litter material. As hypothesised before (Ponsard & Arditi, Reference Ponsard and Arditi2000; Scheu & Falca, Reference Scheu and Falca2000), the postulated trophic level enrichment in δ15N by 3.4‰ does not apply to detritivores. Reviewing 15N enrichment in food chains, Vanderklift & Ponsard (Reference Vanderklift and Ponsard2003) calculated a mean δ15N enrichment of only 0.5 for detritivores. Compared to detritivores, predators were enriched in 15N by on average 3.5‰. This supports our assumption that predators of the animal community studied predominantly feed on prey out of the decomposer system.

Detritus-based food webs are characterized by heterogeneous basal resources; resources in the detritus system originate from the whole trophic spectrum in the food web, including plant and animal residues but also living microorganisms and residues of them. Leaf litter and associated microorganisms, i.e. fungi, bacteria and algae, are important resources of detritivorous arthropods. Since the diet of omnivorous feeders consists of different resources, δ15N signatures of consumers may differ depending on the resource composition of their diet. Trophic level omnivory, i.e. feeding on resources of different trophic levels, is likely to result in high variations in 15N signatures at the group and individual level. In fact, in the present study, δ15N signatures varied strongly within detritivores and predators, which are commonly treated as single trophic levels. Signatures of δ15N of detritivorous arthropods varied by about 8‰ and those of predatory arthropods by about 6‰, indicating different use of food resources among guild members, presumably out of different trophic levels. High variations in δ15N signatures further indicate that species of the trophic guilds do not form homogeneous feeding groups or distinct trophic levels. According to the Scheffé test, the taxa studied consisted of subsets of species with statistically homogeneous δ15N signatures. Trophic groups of detritivores, herbivores, predators and necrovores were distributed over different subsets, indicating the existence of trophic subguilds.

Delineation of trophic subgroups within trophic guilds

Detritivores and predators consisted of species forming a gradient in δ15N signatures. Treated as trophic guilds, detritivores and predators overlapped in their trophic position within the soil animal community. Based on statistically homogeneous groups, seven and four subguilds of detritivores and predators, respectively, were distinguished. Subguilds may represent functional groups within trophic guilds, reflecting gradual differences in resource combinations.

Detritivores

Differing δ15N signatures of detritivores may be due to preferential consumption of habitat specific litter, litter material in different stages of decay or specific litter compartments (Tayasu, Reference Tayasu1998; Scheu & Falca, Reference Scheu and Falca2000; Pollierer et al., Reference Pollierer, Langel, Scheu and Maraun2009). Feeding on habitat specific litter may explain variability in δ15N signatures by about 3 δunits. Litter of the forest was depleted in 15N compared to litter of the meadow and the boundary area (Appendix). Preferential consumption of litter of different stages of decay may contribute to high variance of δ15N signatures among detritivorous species. Compared to fresh litter, decaying litter is enriched in 15N (Nadelhoffer & Fry, Reference Nadelhoffer and Fry1988; Wedin et al., Reference Wedin, Tieszen, Dewey and Pastor1995; Handley & Scrimgeour, Reference Handley and Scrimgeour1997). With progressing decay of litter, the amount of associated microorganisms increases. Microorganisms are known to translocate N into decaying litter (Handley & Scrimgeour, Reference Handley and Scrimgeour1997; Schimel & Hättenschwiler, Reference Schimel and Hättenschwiler2007).

Signatures of δ15N of detritivore subguilds D1–D3 were depleted in 15N compared to mean litter δ15N, and those of detritivore subguild D4 differed little from those of litter. D1–D4 detritivores presumably were limited by nitrogen; however, depletion or low enrichment in 15N may also have resulted from feeding on litter compounds low in 15N. Animals consuming low protein diets recycle rather than excrete nitrogen and synthesize new amino acids out of nitrogen of desaminated proteins (Fisler et al., Reference Fisler, Drenick, Blumfeld and Swendseid1982). Presumably, high nitrogen use efficiency is an adaptation of organisms which consume resources of low nitrogen concentration (Vanderklift & Ponsard, Reference Vanderklift and Ponsard2003). Nitrogen limited organisms have to maximize assimilation of nitrogen in food resources, i.e. decrease nitrogen excretion. Since nitrogen waste products are depleted in 15N compared to animal tissues (Steele & Daniel, Reference Steele and Daniel1978), reduced excretion of nitrogen results in lower 15N fractionation. This applies to phloem-sucking aphids (Ostrom et al., Reference Ostrom, Colunga-Garcia and Gage1997; Yoneyama et al., Reference Yoneyama, Handley, Scrimgeour, Fisher and Raven1997) and, presumably, also to detritivores that rely on fresh litter resources.

Compared to litter, D5 and D6 detritivores were enriched in 15N. D5 and D6 detritivores may represent species, which increasingly gain their energy from microorganisms associated with decaying litter. Digestion of both plant material and microorganisms may facilitate nitrogen uptake by detritivores and, therefore, increase 15N fractionation. Detritivores of subguild D7 were strongly enriched in 15N compared to litter, indicating that they consume high amounts of animal tissue, either as predators or as necrovores. Overall, the results indicate that the postulated 15N discrimination in detritivores depends on nitrogen uptake, which in turn depends on the association of micoroorganisms with litter. Signatures of δ15N of some detritivores corresponded to those of predators. Therefore, detritivores, as commonly defined, may include species which predominantly live on a diet of animal tissue.

In the present study, Collembola dominated detritivorous arthropods. Most Collembola taxa are sapro- and microphageous (Zachariae, Reference Zachariae, Doeksen and van der Drift1963; Wallwork, Reference Wallwork1976; Wolters, Reference Wolters1985; Verhoef et al., Reference Verhoef, Prast and Verweij1988; Chen et al., Reference Chen, Snider and Snider1996; Zettel et al., Reference Zettel, Zettel, Suter, Streich and Egger2002), but their food spectrum also contains other soil arthropods, carcasses, bacteria, fungi and faeces of, for example, Diplopoda (Rusek, Reference Rusek1998), resulting in very different δ15N signatures (Chahartaghi et al., Reference Chahartaghi, Langel, Scheu and Ruess2005).

The Collembola investigated in the present study predominantly live on the soil surface. Signatures of δ15N of Entomobryidae (D1, D2 and D5) varied by 4‰, with Lepidocyrtus sp. having the highest δ15N signature. Signatures of δ15N of each of the Collembola studied resembled that of the litter in their favoured habitat. Since litter of the meadow was enriched in 15N compared to litter of the forest, variances in δ15N signatures within Entomobryidae are likely due to the consumption of habitat specific litter with T. longicornis and O. flavescens feeding on forest litter and Lepidocyrtus sp. feeding on meadow litter.

Diplopoda also are known to feed on litter materials (Striganova, Reference Striganova1967; Blower, Reference Blower1985; Eisenbeis & Wichard, Reference Eisenbeis and Wichard1985; Werner & Dindal, Reference Werner, Dindal, Slansky and Rodriguez1987; Hopkin & Read, Reference Hopkin and Read1992) but spread over four trophic subguilds (D2, D3, D4 and D5). Diplopoda predominantly occurred at the boundary area. Therefore, consumption of litter at later stages of decay, or coprophagy, may have contributed to variances in their δ15N signatures. Signatures of δ15N of Allajulus sp. (D2) were similar to those of forest litter but varied strongly, suggesting a broad food spectrum. Compared to Allajulus sp., Glomeris sp. (D3 and D5) and Macrosternodesmidae (D4) were enriched in 15N. Juvenile stages of Glomeris sp. (D5) were more enriched in 15N than adults (D3) with their δ15N signatures varying only little, suggesting narrow food spectrum. As indicated by 15N enrichment in juvenile Glomeris sp., they presumably rely more on decayed litter material than adults. This might be due to small body size and weaker mandibles since litter tissue becomes softer and enriched with nitrogen with colonization by fungi. This assumption is supported by highest activity density of juvenile Glomeris sp. in June and July when decayed litter from the last autumn predominates.

Isopoda (D5 and D6) consume fresh and decaying litter but also reingest faeces (Striganova, Reference Striganova1967; Dunger, Reference Dunger1983; Eisenbeis & Wichard, Reference Eisenbeis and Wichard1985), which is known to be a common strategy of isopods to improve their nitrogen supply. In the period from May to August, isopods presumably predominantly fed on decomposed leaf litter of the previous year. As in Diplopoda, differences in δ15N signatures of Isopoda may be due to varying amounts of decayed litter or coprophagy.

Low variance in δ15N signatures of Drosophilidae (D6) may be related to feeding on liquids of decaying organic materials. Larvae of Sciaridae (D6) are important decomposers of litter, particularly in forests (Hövemeyer, Reference Hövemeyer1999). Signatures of δ15N of Sciaridae indicate that they consume a mixture of decaying litter and associated microorganisms.

Cecidomyiidae and Phoridae (D7) were enriched in 15N by 6.0‰ compared to litter 15N. Presumably, those detritivores predominantly feed on animal tissue, either as predators or as necrovores. In fact, larvae of Cecidomyiidae have been proposed to live as predators, while adults predominantly feed on fungi (Honomichl, Reference Honomichl1998). High δ15N signatures of Cecidomyiidae, therefore, indicate that the predaceous larval phase determines the δ15N signature of adults. Phoridae are adapted to live in and on the leaf litter layer, with larvae in part living endoparasitic in insects and adults, visiting flowers but in part also feeding on animal prey and carcasses (Honomichl, Reference Honomichl1998). High δ15N signatures of Phoridae indicate them to be mainly necrovorous.

Overall, detritivores presumably consist of three trophic levels with the species forming a gradient from the first to the third: (i) primary decomposers feeding on fresh litter and certain litter compounds; (ii) secondary decomposers predominantly feeding on litter associated microorganisms; and (iii) species predominantly feeding on animal tissue (predators or necrovores).

Predators

As in detritivores, high variance in δ15N signatures in a priori defined predator species indicate marked differences in food resources of predators. Predators did not form a distinct trophic level; rather, they consisted of subguilds of similar δ15N signatures. Most of the predators studied are generalist feeders, hunting on the soil surface. Due to their high abundance, detritivores (81% of the captured individuals) likely formed important prey, which is also indicated by δ15N signatures. Compared to detritivores, nitrogen uptake is more balanced in predators; and, therefore, discrimination of 15N increases at higher trophic levels (Pearson et al., Reference Pearson, Levey, Greenberg and Martinez del Rio2003). Further, 15N fractionation presumably varies little in predators and is conform to the postulated 15N enrichment of 3.4‰ (Minagawa & Wada, Reference Minagawa and Wada1984; Post, Reference Post2002). Variances in δ15N signatures of predators may be due to preferential consumption of a specific subguild of detritivores, to different amounts of intraguild prey or plant material in the food spectrum.

Predators consisted of four trophically homogeneous subguilds. Presumably, predators P1 predominantly feed on primary decomposers, as, compared to these, they were enriched in 15N by 3.4‰. Signatures of δ15N of predators P2 indicate that they may consume mainly secondary decomposers (D4, ▵=3.8‰), which rely on microorganisms associated with litter material and/or herbivores (H2, ▵=3.7‰ and H3, ▵=3.5‰). As indicated by low δ15N signatures intraguild predation and cannibalism are likely of minor importance. The prey spectrum of predators P3 presumably consists in large part of secondary decomposers (D6, ▵=3.7‰). Intraguild predation and cannibalism likely becomes increasingly important from predators P3 to P4. Predators P4 may consume predominantly intraguild prey (P2, ▵=3.1‰). Overall, the predator community studied appear to consist of trophic subguilds differing in the relative contribution of detritivores, herbivores and predators to their food spectrum.

Most of the spiders studied are mobile hunters feeding in both the meadow and forest. Spiders of the study site were distributed among different predator subguilds; they belonged to predators P2–P4. Signatures of δ15N of P2 spiders indicate that intraguild predation is of little importance in these species. Z. spinimana hunts in the leaf litter layer and, likely, predominantly feeds on detritivores, such as Collembola. Supporting this suggestion, Z. spinimana was enriched in 15N by 3.4‰ compared to T. longicornis and by 4.1‰ compared to O. flavescens. E. frontalis hunts in lower vegetation and O. praticola in higher vegetation (Roberts, Reference Roberts1995), suggesting they feed on herbivores. Indeed, both spider species were enriched in 15N by ca. 3‰ compared to Muellerianella sp. and J. pseudocellaris, which were the dominating herbivores at the sampling site.

Results of our study suggest that differences in δ15N signatures of spiders are not related to hunting strategies. Agelenid spiders use webs on the soil surface to catch their prey and, consequently, are tied to the place of their web; in contrast, lycosid and gnaphosid spiders actively hunt on the soil surface. Thomisid and pisaurid spiders hunt actively in the lower vegetation (Roberts, Reference Roberts1995). Despite different hunting strategies of agelenid and lycosid spiders, δ15N signatures of P. lugubris and H. torpida were similar, indicating that the prey spectra of these spiders overlap; and, therefore, they may compete for prey. Although hunting strategies of lycosid and gnaphosid spiders are similar, δ15N signatures differed significantly, indicating differences in the relative contribution of detritivores, herbivores and predators to their food spectrum. Interestingly, δ15N signatures also differed between developmental stages of lycosid species, indicating changes in the food spectrum with age and body size.

Consistent with the assumption that the trophic position of predators scales with body size, δ15N signature was at a maximum in D. fimbriatus, the biggest spider studied, suggesting that this species was the most vigorous intraguild predator. In a closely related species, D. triton, every developmental stage is known to be cannibalistic (Zimmermann & Spence, Reference Zimmermann and Spence1989). However, the trophic position of spiders did not scale uniformly with body size, e.g. δ15N signature of the small species P. degeeri also was high, suggesting that intraguild predation also is important in smaller species.

As in spiders, δ15N signatures of carabid beetles varied strongly, suggesting that they feed on very different prey. Larvae and adults of carabid beetles are predominantly predaceous. N. biguttatus, Amara sp. and Leistus sp. showed the postulated trophic level enrichment in 15N compared to the Collembola species studied, indicating that they predominantly feed on Collembola. Supporting this suggestion, N. biguttatus hunts predominantly Collembola, including surface living species such as Orchesella cincta and Tomocerus minor (Ernsting, Reference Ernsting1977; Ernsting et al., Reference Ernsting, Isaaks and Berg1992). Sunderland (Reference Sunderland1975) demonstrated, that Collembola contribute 78% to the total prey of N. biguttatus. Also, Leistus sp. is known to predominantly feed on Collembola (Honomichl, Reference Honomichl1998). Some of the carabid species studied, such as Pterostichus spp., Harpalus spp. and Carabus spp., are known to also live on plant resources (Sunderland, Reference Sunderland1975; Honomichl, Reference Honomichl1998). As indicated by the high intraspecific variation in δ15N signatures, especially Harpalus spp. may regularly consume plant materials. Poecilus sp. and D. globosus were the carabid beetles with the highest δ15N signatures. Dyschirius species are very small, live in the soil and consume predominantly staphylinid beetles and Heteroceridae (Eisenbeis & Wichard, Reference Eisenbeis and Wichard1985). D. globosus preferably consumes Enchytraeidae (Honomichl, Reference Honomichl1998), which were not analysed in this study but have been shown to be rather enriched in 15N (by ∼4‰ compared to plant residues: Albers et al., Reference Albers, Schaefer and Scheu2006).

Despite sharing the same habitat, adult staphylinid beetles (P2) were depleted in 15N compared to larvae (P3), indicating that intraguild predation is more important in larvae than in adults. Signatures of δ15N of the adult staphylinid beetle species studied were similar, indicating consumption of prey of similar trophic position. In some staphylinid beetles, Collembola and aphids form important prey as, for example, in Stenus spp. (Sunderland et al., Reference Sunderland, Crook, Stacey and Fuller1987; Honomichl, Reference Honomichl1998). In our study, Stenus spp. was enriched in 15N by ca. 5.7‰ compared to Collembola, suggesting that they also consume detritivores of higher trophic position and/or intraguild prey.

The analysis of stable isotopes of nitrogen is a powerful tool to depict the structure of food webs predominated by generalist feeders (Schmidt et al., Reference Schmidt, Scrimgeour and Handley1997; Tayasu et al., Reference Tayasu, Abe, Eggleton and Bignell1997; Neilson et al., Reference Neilson, Hamilton, Wishart, Marriotti, Boag, Handley, Scrimgeour, McNicol and Robinson1998; Briones et al., Reference Briones, Ineson and Sleep1999; McNabb et al., Reference McNabb, Halaj and Wise2001; Oelbermann et al., Reference Oelbermann, Langel and Scheu2008). Further, the method, as applied in the present study, may allow depicting interactions within trophic levels. However, until today, only little was known about the pattern of 15N enrichment in detritus-based soil animal communities; therefore, the analysis of variances in natural stable isotopes of nitrogen needs to be interpreted with caution. As results of the present study indicate, the postulated stepwise enrichment in 15N by 3.4‰ per trophic level (Minagawa & Wada, Reference Minagawa and Wada1984) is not universal, i.e. detritivores are likely to deviate from this rule. Results further indicate that the postulated 15N discrimination in detritivores depends on nitrogen uptake, which in turn depends on the association of microorganisms with litter.

Overall, the results indicate that both species within trophic levels and species within taxonomic groups consist of trophic subguilds differing in food spectrum. Trophic differentiation is most pronounced in detritivores which comprise subguilds predominantly feeding on certain litter compartments, litter of different stages of decay and animal tissue. Taking them as trophic species or trophic levels, as commonly done in food web studies (Moore & De Ruiter, Reference Moore and De Ruiter1991; De Ruiter et al., Reference De Ruiter, Neutel, Moore, Polis and Winemiller1996), therefore, is inappropriate. To develop strategies for improving the control of herbivore pest species by generalist predators, it is particularly important to identify predators which consume both decomposers and herbivores. Rather than ascribing species to fixed trophic levels, a more detailed delineation of trophically homogeneous groups is necessary for understanding food web links and interactions.

Acknowledgements

The work was funded by the Deutsche Forschungsgemeinschaft.

Appendix

Species composition of statistical homogeneous subgroups, their assignment to trophic subguilds (B, base of the food web; H1–H3, herbivores; D1–D7, detritivores; P1–P4, predators and N1–N3, necrovores), their δ15N signatures (±SD) and main habitat.

References

Albers, D., Schaefer, M. & Scheu, S. (2006) Incorporation of plant carbon into the soil animal food web of an arable system. Ecology 87, 235–245.Google Scholar
Bengtsson, J., Lundkvist, H., Saetre, P., Sohlenius, B. & Solbreck, B. (1998) Effects of organic matter removal on the soil food web: forestry practices meet ecological theory. Applied Soil Ecology 9, 137–143.CrossRefGoogle Scholar
Blower, J.G. (1985) Millipedes. Synopses of the British Fauna (New Series) 35. London, Brill Backhuys.Google Scholar
Briones, M.J.I., Ineson, P. & Sleep, D. (1999) Use of δ13C to determine food selection in collembolan species. Soil Biology and Biochemistry 31, 937–940.Google Scholar
Chahartaghi, M., Langel, R., Scheu, S. & Ruess, L. (2005) Feeding guilds in Collembola based on nitrogen stable isotope ratios. Soil Biology and Biochemistry 37, 1718–1725.CrossRefGoogle Scholar
Chen, B., Snider, R.J. & Snider, R.M. (1996) Food consumption by Collembola from northern Michigan deciduous forest. Pedobiologia 40, 149–161.CrossRefGoogle Scholar
Chinery, M. (1993) Parey's Buch der Insekten. Hamburg and Berlin, Germany, Verlag Paul Parey.Google Scholar
Coleman, D.C. & Crossley, D.A. (1996) Fundamentals of Soil Ecology. San Diego, California, USA, Academic Press.Google Scholar
De Ruiter, P.C., Neutel, A.M. & Moore, J.C. (1996) Energetics and stability in below-ground food webs. pp. 201–210in Polis, G.A. & Winemiller, K.O. (Eds) Food Webs. Integration of Patterns and Dynamics. New York, USA, Chapman and Hall.Google Scholar
Dücker, A., Schmüser, H., Heubel, K., Borcherding, R., Heubel, V., Müller-Reich, C., Pahnke, K., Gienapp, P., Nötzold, R. & Nötzold, V. (1997) Laufkäfer. Hamburg, Germany, Deutscher Jugendbund für Naturbeobachtung.Google Scholar
Dunger, W. (1983) Tiere im Boden. 3. Auflage. Neue Brehm-Bücherei. A. Wittenberg Lutherstadt, Germany, Ziemsen-Verlag.Google Scholar
Eisenbeis, G. & Wichard, W. (1985) Atlas zur Biologie der Bodenarthropoden. Stuttgart, Germany, Gustav Fischer Verlag.Google Scholar
Ernsting, G. (1977) Aspects of predation and the coexistence of collembolan prey species. Ecological Bulletins (Stockholm) 25, 478–480.Google Scholar
Ernsting, G., Isaaks, J.A. & Berg, M.P. (1992) Life cycle and food availability indices in Notiophilus biguttatus (Coleoptera, Carabidae). Ecological Entomology 17, 33–42.CrossRefGoogle Scholar
Fisler, J.S., Drenick, E.J., Blumfeld, D.E. & Swendseid, M.E. (1982) Nitrogen economy during very low calorie and very low protein diets. American Journal of Clinical Nutrition 35, 471–486.Google Scholar
Fjellberg, A. (1980) Identification keys to the Norwegian Collembola. Ås, Norway, Norsk Entomologisk Forening.Google Scholar
Freude, H., Harde, K.W. & Lohse, G.A. (1976) Die Käfer Mitteleuropas, Band 2 Adephaga 1. Krefeld, Germany, Goecke & Evers.Google Scholar
Gannes, L.Z., O'Brien, D.M. & Martinez del Rio, C. (1997) Stable isotopes in animal ecology: assumptions, caveats, and a call for more laboratory experiments. Ecology 78, 1271–1276.CrossRefGoogle Scholar
Hairston, N.G. & Hairston, N.G. (1993) Cause-effect relationships in energy flow trophic structure, and interspecific interactions. The American Naturalist 142, 379–411.Google Scholar
Halaj, J. & Wise, D.H. (2002) Impact of detrital subsidy on trophic cascades in a terrestrial grazing food web. Ecology 83, 3141–3151.CrossRefGoogle Scholar
Handley, L.L. & Scrimgeour, C.M. (1997) Terrestrial plant ecology and 15-N natural abundance: the present limits to interpredation for uncultivated systems with original data from a Scottish old field. Advances in Ecological Research 27, 139–212.Google Scholar
Harde, K.W. & Severa, F. (1998) Der Kosmos-Käferführer. Stuttgart, Germany, Franckh'sche Verlagsbuchhandlung.Google Scholar
Heimer, S. & Nentwig, W. (1991) Spinnen Mitteleuropas. Berlin, Germany, Paul Parey.Google Scholar
Honomichl, K. (1998) Jacobs/Renner-Biologie und Ökologie der Insekten. 3.Auflage. Stuttgart, Germany, Gustav Fischer.Google Scholar
Hopkin, S.P. & Read, H.J. (1992) The Biology of Millipedes. New York, USA, Oxford University Press.Google Scholar
Hövemeyer, K. (1999) Abundance patterns in terrestrial dipteran communities. Pedobiologia 43, 28–43.CrossRefGoogle Scholar
Hunt, H.W., Coleman, D.C., Ingham, E.R., Inham, R.E., Elliott, E.T., Moore, J.C., Rose, S.L., Reid, C.P.P. & Morley, C.R. (1987) The detrital food web in a shortgrass prairie. Biology and Fertility of Soils 3, 57–68.Google Scholar
Jones, D. (1990) Der Kosmos-Spinnenführer. Mitteleuropäische Spinnen und Weberknechte. Stuttgart, Germany, Kosmos-Franckh'sche Verlagsbuchhandlung.Google Scholar
Locket, G.H. & Millidge, A.F. (1951) British Spiders, Vol. I, II. London, UK, Ray Society.Google Scholar
McNabb, D.M., Halaj, J. & Wise, D.H. (2001) Inferring trophic positions of generalist predators and their linkage to the detrital food web in agroecosystems: a stable isotope analysis. Pedobiologia 45, 289–297.Google Scholar
Minagawa, M. & Wada, E. (1984) Stepwise enrichment of 15N along food chains: Further evidence and the relation between Delta 15N and animal age. Geochimica et Cosmochimica Acta 48, 1135–1140.CrossRefGoogle Scholar
Moore, J.C. & De Ruiter, P.C. (1991) Temporal and spatial heterogeneity of trophic interactions within below-ground food webs. Agriculture Ecosystems and Environment 34, 371–397.Google Scholar
Nadelhoffer, K.J. & Fry, B. (1988) Controls of nitrogen-15 and carbon-13 abundances in forest soil organic matter. Soil Science Society of America Journal 52, 1633–1640.CrossRefGoogle Scholar
Neilson, R., Hamilton, D., Wishart, J., Marriotti, C.A., Boag, L., Handley, L.L., Scrimgeour, C.M., McNicol, J.W. & Robinson, D. (1998) Stable isotope natural abundances of soil, plants and soil invertebrates in an upland pasture. Soil Biology and Biochemistry 30, 1773–1782.Google Scholar
Nyffeler, M., Sterlling, W.L. & Dean, D.A. (1994) How spiders make a living. Environmental Entomology 23, 1357–1367.CrossRefGoogle Scholar
Oelbermann, K., Langel, R. & Scheu, S. (2008) Utilization of prey from the decomposer system by predators of grassland. Oecologia 155, 605–617.CrossRefGoogle ScholarPubMed
Ostrom, P.H., Colunga-Garcia, M. & Gage, S.H. (1997) Establishing pathways of energy flow for insect predators using stable isotope ratios: field and laboratory evidence. Oecologia 109, 108–113.Google Scholar
Peterson, B.J. & Fry, B. (1987) Stable isotopes in ecosystem studies. Annual Review of Ecology and Systematics 18, 293–320.Google Scholar
Pearson, S.F., Levey, D.J., Greenberg, C.H. & Martinez del Rio, C. (2003) Effects of elemental composition on the incorporation of dietary nitrogen and carbon isotopic signatures in an omnivorous songbird. Oecologia 135, 516–523.CrossRefGoogle Scholar
Polis, G.A. (1991) Complex trophic interactions in deserts: an empirical critique of food-web theory. The American Naturalist 138, 123–155.Google Scholar
Polis, G.A. & Strong, D.R. (1996) Food web complexity and community dynamics. The American Naturalist 147, 813–846.Google Scholar
Polis, G.A., Myers, C.A. & Holt, R.D. (1989) The ecology and evolution of intraguild predation: potential competitors that eat each other. Annual Reviews of Ecology and Systematics 20, 297–330.Google Scholar
Pollierer, M., Langel, R., Scheu, S. & Maraun, M. (2009) Compartmentalization of the soil animal food web as indicated by dual analysis of stable isotope ratios (15N/14N and 13C/12C). Soil Biology and Biochemistry 41, 1221–1226.CrossRefGoogle Scholar
Ponsard, S. & Arditi, R. (2000) What can stable isotopes (delta 15N and delta 13C) tell about the food web of soil macro-invertebrates? Ecology 81, 852–864.Google Scholar
Post, D.M. (2002) Using stable isotopes to estimate trophic position: models, methods and assumptions. Ecology 83, 703–718.Google Scholar
Reineking, A., Langel, R. & Schikowski, J. (1993) 15N, 13C-on-line-measurements with an elemental analyser (Carlo Erba, NA 1500), a modified trapping box and a gas isotope mass spectrometer (Finnigan, MAT 251). Isotopenpraxis Environmental Health Studies 29, 169–174.CrossRefGoogle Scholar
Roberts, M.J. (1995) Collins Field Guide. Spiders of Britain and Northern Europe. London, UK, HarperCollins Publishers.Google Scholar
Rusek, J. (1998) Biodiversity of Collembola and their functional role in the ecosystem. Biodiversity and Conservation 7, 1207–1219.CrossRefGoogle Scholar
Sauer, F. (1996) Wanzen und Zikaden. Karlsfeld, Germany, Fauna-Verlag.Google Scholar
Sauer, F. (1998) Fliegen und Mücken. Karlsfeld, Germany, Fauna-Verlag.Google Scholar
Schaefer, M. (1990) The soil fauna of a beech forest on limestone: trophic structure and energy budget. Oecologia 82, 128–136.Google Scholar
Schaefer, M. (1991) The animal community: diversity and resources. pp. 51–120 in Röhrig, E. & Ulrich, B. (Eds) Temperate Deciduous Forests. Ecosystems of the World. Amsterdam, The Netherlands, Elsevier.Google Scholar
Schaefer, M. (1992) Brohmer – Fauna von Deutschland. Heidelberg, Germany, Quelle & Meyer.Google Scholar
Scheu, S. (2002) The soil food web: structure and perspectives. European Journal of Soil Biology 38, 11–20.CrossRefGoogle Scholar
Scheu, S. & Falca, M. (2000) The soil food web of two beech forests (Fagus sylvatica) of contrasting humus type: stable isotope analysis of a macro- and a mesofauna-dominated community. Oecologia 123, 285–296.Google Scholar
Scheu, S. & Setälä, H. (2002) Multitrophic interactions in decomposer food webs. pp. 223–264in Tscharntke, T. & Hawkins, B.A. (Eds) Multitrophic Level Interactions. Cambridge, UK, Cambridge University Press.CrossRefGoogle Scholar
Schimel, J.P. & Hättenschwiler, S. (2007) Nitrogen transfer between decomposing leaves of different N status. Soil Biology and Biochemistry 39, 1428–1436.CrossRefGoogle Scholar
Schmidt, O., Scrimgeour, C.M. & Handley, L.L. (1997) Natural abundance of 15N and 13C in earthworms from a wheat and a wheat-clover field. Soil Biology and Biochemistry 29, 1301–1308.Google Scholar
Schmidt, O., Curry, J.P., Dyckmans, J., Rota, E. & Scrimgeour, C.M. (2004) Dual stable isotope analysis (d13C and d15N) of soil invertebrates and their food sources. Pedobiologia 48, 171–180.Google Scholar
Steele, K.W. & Daniel, R.M. (1978) Fractionation of nitrogen isotopes by animals: a further complication to the use of variations in the natural abundance of 15N for tracer studies. Journal of Agricultural Science 90, 7–9.Google Scholar
Striganova, B.R. (1967) Über die Zersetzung von überwinterter Laubstreu durch Tausendfüßer und Landasseln. Pedobiologia 7, 125–138.Google Scholar
Sunderland, K.D. (1975) The diet of some predatory arthropods in cereal crops. Journal of Applied Ecology 12, 507–515.Google Scholar
Sunderland, K.D., Crook, N.E., Stacey, D.L. & Fuller, B.J. (1987) A study of feeding by polyphagous predators on cereal aphids using ELISA and gut dissection. Journal of Applied Ecology 24, 907–933.Google Scholar
Swift, M.F., Heal, O.W. & Anderson, J.M. (1979) Decomposition in terrestrial ecosystems. Berkeley, California, USA, University of California Press.CrossRefGoogle Scholar
Tayasu, I. (1998) Use of carbon and nitrogen isotope ratios in termite research. Ecological Research 13, 377–387.Google Scholar
Tayasu, I., Abe, T., Eggleton, P. & Bignell, D.E. (1997) Nitrogen and carbon isotope ratios in termites – an indicator of trophic habit along the gradient from wood-feeding to soil-feeding. Ecological Entomology 22, 343–351.CrossRefGoogle Scholar
Vanderklift, M.A. & Ponsard, S. (2003) Sources of variation in consumer-diet δ15N enrichment: a meta-analysis. Oecologia 136, 169–182.CrossRefGoogle ScholarPubMed
Verhoef, H.A., Prast, J.E. & Verweij, R.A. (1988) Relative importance of fungi and algae in the diet and nitrogen nutrition of Orchesella cincta (L.) and Tomocerus minor (Lubbock) (Collembola). Functional Ecology 2, 195–201.Google Scholar
Wachmann, E., Platen, R. & Barndt, D. (1995) Laufkäfer: Beobachtung und Lebensweise. Augsburg, Germany, Naturbuch-Verlag.Google Scholar
Wallwork, J.A. (1976) The Distribution and Diversity of Soil Fauna. London, UK, Academic Press.Google Scholar
Wedin, D.A., Tieszen, L.L., Dewey, B. & Pastor, J. (1995) Carbon isotope dynamics during grass decomposition and soil organic matter formation. Ecology 76, 1383–1392.CrossRefGoogle Scholar
Werner, M.R. & Dindal, D.L. (1987) Nutritional Ecology of soil arthropods. In: Nutritional ecology of insects, mites, spiders and related invertebrates. pp. 815–836in Slansky, F. & Rodriguez, J.G. (Eds) Nutritional Ecology of Insects, Mites, and Spiders. New York, USA, J. Wiley.Google Scholar
Witt, R. (1998) Wespen – beobachten, bestimmen. Augsburg, Germany, Naturbuch-Verlag.Google Scholar
Wolters, V. (1985) Untersuchung zur Habitatbindung und Nahrungsbiologie der Springschwänze (Collembola) eines Laubwaldes unter besonderer Berücksichtigung ihrer Funktion in der Zersetzerkette. PhD dissertation, University of Göttingen, Göttingen, Germany.Google Scholar
Yoneyama, T., Handley, L.L., Scrimgeour, C.M., Fisher, D.B. & Raven, J.A. (1997) Variations of the natural abundances of nitrogen and carbon isotopes in Triticum aestivum, with special reference to phloem and xylem exudates. New Phytologist 137, 205–213.CrossRefGoogle ScholarPubMed
Zachariae, G. (1963) Was leisten Collembolen für den Waldhumus? pp. 109–124in Doeksen, J. & van der Drift, J. (Eds). Proceedings of the Colloquium on Soil Fauna, Soil Microflora and their Relationships. Amsterdam, The Netherlands, North-Holland Publishing Company.Google Scholar
Zahradnik, J. (1985) Käfer Mittel- und Nordwesteuropas. Hamburg, Germany, Paul Parey.Google Scholar
Zettel, J., Zettel, U., Suter, C., Streich, S. & Egger, B. (2002) Winter feeding behaviour of Ceratophysella sigillata (Collembola: Hypogastruridae) and the significance of eversible vesicles for resource utilisation. Pedobiologia 46, 404–413.Google Scholar
Zheng, D.W., Bengtsson, J. & Agren, G.I. (1997) Soil food webs and ecosystem processes-decomposition in donor-control and Lotka-Volterra systems. American Naturalist 149, 125–148.CrossRefGoogle Scholar
Zimmermann, M. & Spence, J.R. (1989) Prey use of the fishing spider Dolomedes triton (Pisauridae, Araneae): An important predator of the neuston community. Oecologia 80, 187–194.CrossRefGoogle ScholarPubMed
Figure 0

Fig. 1. Mean δ15N signatures (±SD) of litter (L), herbivores (H), detritivores (D), predators (P) and necrovores (N) in a forest-meadow transect. n, the number of species representing this group. Significant differences are represented by different letters (Tukey's HSD test, P<0.05). For aggregation of taxa, see Appendix.