Oecologia (2004) 139: 336–346 DOI 10.1007/s00442-004-1514-6 ECOPHYSIOLOGY Liliane Ruess . Max M. Häggblom . Reinhard Langel . Stefan Scheu Nitrogen isotope ratios and fatty acid composition as indicators of animal diets in belowground systems Received: 28 October 2003 / Accepted: 21 January 2004 / Published online: 6 March 2004 # Springer-Verlag 2004 Abstract This study analyses trophic interactions be- tween soil fungi, micro- and mesofauna in microcosm experiments. The trophic shift of 15 N and fatty acids (FAs) was investigated in different food chains, which comprised either two (fungi and grazers) or three (fungi, nematodes and Collembola) levels. Contrary to the widely accepted assumption of 15 N enrichment in trophic cascades the experiments revealed enrichment, depletion or no change in 15 N of consumers compared to their diet. Factors responsible for this pattern were suggested to be: (1) the main metabolic pathway used for N excretion in ammo- notelic nematodes to be similar or depleted in the heavier isotope, and uricotelic Collembola mostly enriched in the heavier isotope; (2) a higher shift in 15 N with a high- protein diet (e.g. for predators); (3) compensation due to low-quality food altering the fractionation of 15 N. Analysis of the lipid composition showed phospholipids to be generally unaffected and neutral lipids closely related to the FA pattern of the food source. Dietary routing of FAs into neutral lipids occurred, as evidenced by correspond- ing frequencies of FAs in host and consumer profiles. Additionally, several FAs were only detected in the grazer when present in the food source. Oleic acid showed a shift over three trophic levels, from fungi to nematodes to Collembola. The assimilation of dietary FAs resulted in a more diverse neutral lipid profile, i.e. animals higher in the food chain contained more individual FAs compared to animals lower in the food chain. The results indicate that monoenoic C18 and monoenoic C20 FAs have the potential to act as tools for the bioindication of feeding strategies in belowground systems. We suggest that primary consumers will have no or only trace amounts of monoenoic C20 acids in their neutral lipid profile, whereas consumers feeding on a eukaryote diet will show a considerably higher frequency. Keywords Soil fauna . Fungi . Trophic transfer . Lipids . Stable isotopes Introduction Animal diets are usually verified by an analysis of gut contents and faeces or by observations on their foraging. However, these data are restrictive because they only give information on food consumed during a brief window of time. Additionally, there may be a discrepancy between diet incorporation and assimilation. The use of these methods is particularly limited in belowground systems, due to the small size of the fauna and their cryptic habitat. Assigned feeding guilds in soil environments therefore often reflect more taxonomic, rather than functional, relationships. Over the last decade, stable isotope analyses have increasingly been used to determine the relative contribution of different food sources to a consumer ’ s diet (DeNiro and Epstein 1981; Gannes et al. 1997, 1998; Ponsard and Arditi 2000; Post 2002). Isotopic ratios of δ 15 N and δ 13 C in animal tissues serve as naturally occurring labels from which to reconstruct diets, with N isotopes frequently applied to rank animals according to their relative trophic levels. As the δ 15 N values of the whole bodies of animals are usually more positive than those of their diets, there is a stepwise enrichment of δ 15 N along the food chain (Minagawa and Wada 1984; Wada et al. 1991; Eggers and Jones 2000). Stable isotopes provide a powerful tool in food web analysis, but the use of this method relies on the general L. Ruess (*) . S. Scheu Institut für Zoologie, Technische Universität Darmstadt, Schnittspahnstrasse 3, 64287 Darmstadt, Germany e-mail: ruess@bio.tu-darmstadt.de M. M. Häggblom Department of Biochemistry and Microbiology, Cook College, Rutgers University, 76 Lipman Drive, New Brunswick, NJ, 08901–8525, USA R. Langel Kompetenzzentrum Stabile Isotope, Forschungszentrum Waldökosysteme, Georg-August Universität Göttingen, Büsgenweg 2, 37077 Göttingen, Germany