Grazer–diatom size relationships in an epiphytic community

We investigated the animal–food size relationships of an invertebrate grazer community that feeds on the epiphytic diatoms of mosses in a Québec stream. The relationship between invertebrate head width and maximum size of ingested diatoms was significant (r2 5 0.57, n 5 82, p , 0.00001). This relationship did not differ significantly among widely different taxa of insects, crustaceans, and worms. The comparison of our results with relationships developed in the laboratory for zooplankton and benthos indicated that artificial particles or simplified algal assemblages underestimate the size range of algae ingested by grazers. The examination of the size distribution of the ingested diatoms and of those available in the environment showed that grazers fed preferentially on different diatom size classes in accordance with their head width. Concepts of resource partitioning, optimal foraging, and size efficiency proposed for other habitat types are relevant to stream invertebrate communities. This recognition should facilitate the study of algal–grazer interactions in streams. The relationship between the size of an animal and the food it ingests has long been a central topic in ecology. Several predation and competition theories, such as niche, optimal foraging, and resource partitioning, often explicitly quantify food in terms of its size. Animal–food size relationships have been studied in terrestrial as well as aquatic systems (Burns 1968; Vézina 1985). In planktonic communities, predator–prey size relationships have been demonstrated for piscivorous and planktivorous fish and for zooplankton. These relationships form the basis of the size efficiency hypothesis that has been so influential in theoretical and applied limnology (Brooks and Dodson 1965). The surprising regularities in plankton size distributions observed across several orders of magnitude in oceans and lakes are generally attributed to a consistent animal–food size relationship between trophic levels (Sheldon et al. 1972). Such allometric relationships have received little attention in the context of benthic communities. An application of theories elaborated for planktonic communities might facilitate the study of the notoriously complex benthic communities. Previous studies of benthic invertebrate herbivory have focused mostly on one or a few populations of large grazers that are easily manipulated in the laboratory or, more rarely, in natural streams (see review in Steinman 1996). Such studies show the effect of these grazers on periphyton but provide little information about natural communities in which invertebrates range widely in size and taxonomy. Benthic grazers have been traditionally considered generalists rather than specialist feeders (Cummins 1973; Mihuc 1997), but some evidence suggests that they might select algae on the basis of size, life form, chemical composition, or a combination of factors (Peterson 1987; Steinman 1996; Hamels et al. 2004). In this study, we have analyzed the size range of epiphytic diatoms consumed by a grazer community associated with aquatic moss in a Québec stream. The technique we used to examine their gut contents involved the oxidation of organic matter, which allowed the analysis of particles ingested by small grazers without dissection. This technique, however, limits observations to diatoms. For this reason, we chose an epiphytic community in which diatoms represented .90% of total algal density and biomass. In this paper, we assess the application of ecological theories on predator–prey size to a benthic system. In particular, we test whether (1) there is a relationship between the size of ingested diatoms and grazer size, (2) this relation differs according to grazer taxon, and (3) grazers feed selectively on diatoms as a function of their size. We then compare our results to grazer–prey size relationships observed for planktonic communities. Materials and methods—Diatoms and grazers were collected from a second-order stream, Ruisseau Pin Rouge, which is 80 km north of Montréal in the Laurentians Mountains (458599N, 74819W). This stream is oligomesotrophic with slightly acid, colored, low-conductivity water (Cattaneo et al. 2004). The aquatic moss Fontinalis dalecaerlica grows profusely on the side of large boulders, where it forms a habitat for periphyton and invertebrates. This study is based on quantitative samples of moss with their epiphytic algae and invertebrates that were collected along a 100-m-long run in May 1998 to study the effects of varying discharge on stream communities (Cattaneo et al. 2004). We analyzed the gut content of invertebrates belonging to the main grazer taxa found in the moss. Chironomids were the most abundant grazers (Table 1). Individuals were macerated in concave slides with 30% hydrogen peroxide for 3–24 h depending on size and cuticle texture and then mounted in Hoyer’s mounting medium (Humason 1967). After mounting, grazers were identified to genus (Wiederholm 1983; Merrit and Cummins 1996). The maceration procedure enabled observation of the diatoms contained in the gut. We counted and identified the ingested diatoms to genus or species (Table 2) and then measured their greatest axial length dimension (GALD) to the nearest 1 mm under a microscope (3400) connected to an image analysis system (Image Pro Plus, version 4.1 for Windows, Media Cybernetics). When .100 diatoms were present in a gut, we analyzed a subsample. Only grazers with at least 10 diatoms present in the gut were used in the analysis. We chose head width as a de-