About 10 years ago, reviews of the use of stable isotopes in animal ecology predicted explosive growth in this field and called for laboratory experiments to provide a mechanistic foundation to this growth. They identified four major areas of inquiry: (1) the dynamics of isotopic incorporation, (2) mixing models, (3) the problem of routing, and (4) trophic discrimination factors. Because these areas remain central to isotopic ecology, we use them as organising foci to review the experimental results that isotopic ecologists have collected in the intervening 10 years since the call for laboratory experiments. We also review the models that have been built to explain and organise experimental results in these areas.

Summary Ten years ago Gannes et al. (1997, Stable isotopes in animal ecology: assumptions, caveats, and a call for laboratory experiments. Ecology , 78 , 1271-1276, 1998) identified four major areas requiring further research in experimental animal isotopic ecology: (i) the dynamics of isotopic incorporation, (ii) mixing models, (iii) the problem of routing, and (iv) trophic discrimination factors. 2. Differences in isotopic incorporation rates among tissues seem to be explained by variation in protein turnover. The application of multi-compartment models to isotopic incorporation data has revealed that different inferences can be derived between these and one-compartment models. 3. A variety of mixing models of varying degrees of complexity and realism are used to find the contribution of isotopic sources to the elements in an organism's tissues. The use of these models demands the use of tissue to diet discrimination factors that are rarely measured experimentally. 4. Mixing models assume that assimilated nutrients are disassembled into their elemental components and that these elements are reassembled into biomolecules. This assumption is unrealistic as macromolecules are routed differentially into tissues. Isotopic routing is an area that isotopic ecologists have neglected in their experimental and modelling research. 5. Isotopic ecologists are just beginning to understand why 15 N biomagnifies along trophic chains, and to explore the factors that determine the degree of 15 N biomagnification. We review the hypotheses that explain why 15 N biomagnifies up trophic chains. 6. The use of compound-specific isotopic analyses is opening new fruitful areas of research at the intersection of nutritional and isotopic ecology.

Stable-isotope analysis (SIA) has revolutionized animal ecology by providing time-integrated estimates of the use of resources and/or habitats. SIA is based on the premise that the isotopic composition of a consumer's tissues originates from its food, but is offset by trophic-discrimination (enrichment) factors controlled by metabolic processes associated with the assimilation of nutrients and the biosynthesis of tissues. Laboratory preparation protocols dictate that tissues both of consumers and of their potential prey be lipid-extracted prior to analysis, because (1) lipids have carbon isotope (δ(13)C) values that are lower by approximately 3-8‰ than associated proteins and (2) amino acids in consumers' proteinaceous tissues are assumed to be completely routed from dietary protein. In contrast, models of stable-isotope mixing assume that dietary macromolecules are broken into their elemental constituents from which non-essential amino acids are resynthesized to build tissues. Here, we show that carbon from non-protein dietary macromolecules, namely lipids, was used to synthesize muscle tissue in an omnivorous rodent (Mus musculus). We traced the influence of dietary lipids on the synthesis of consumers' tissues by inversely varying the dietary proportion of C4-based lipids and C3-based protein while keeping carbohydrate content constant in four dietary treatments, and analyzing the δ(13)C values of amino acids in mouse muscle after 4 months of feeding. The influence of dietary lipids on non-essential amino acids varied as function of biosynthetic pathway. The source of carbon in ketogenic amino acids synthesized through the Krebs cycle was highly sensitive to dietary lipid content, with significant increases of approximately 2-4‰ in Glutamate and Aspartate δ(13)C values from the 5% to 15% dietary lipid treatment. Glucogenic amino acids (Glycine and Serine) were less sensitive to dietary lipid, but increased by approximately 3-4‰ from the 25% to 40% lipid diet. As expected, the δ(13)C values of essential amino acids did not vary significantly among diets. Although lipids provide a calorie-rich resource that fuels energy requirements, our results show that they also can be an important elemental source of carbon that contributes to the non-essential amino acids used to build structural tissue like muscle. As such, the calculation of trophic-discrimination factors for animals that consume a lipid-rich diet should consider lipid carbon as a building block for proteinaceous tissues. Careful consideration of the macromolecular composition in the diet of the consumer of interest will help to further refine the use of SIA to study animal ecology and physiology.