This becomes increasingly difficult with decreasing size of the food items, because experience shows that the effort required for identification is inversely related to organism size. This is especially true in unicellular organisms such as autorophic or heterotrophic eukaryonts (e.g. algae and ciliates) and even more in procaryonts.

It is, e.g., extremely laborious to sort the "plankton" ingested by a filter feeder or the "microorganisms" a deposit feeder depends on into species. In those extreme cases we may be glad to know the preferred food particle size spectrum (e.g. Vahl 1973) or whether cetain higher taxa such as bacteria are used at all (e.g. Grossmann & Reichardt 1991).

Problems may arise, too, when working in systems with many species and/or incomplete taxonomic records such as the Deep Sea or certain tropical regions. Regarding taxonomic groups we are no experts in, we may be forced to retreat to higher taxonomic identification levels such as genus or family.

Hence trophic source identification is always a tradeoff between resolution obtained and effort invested.