Predicting the outcome of trophic manipulation in lakes—a comment on Harris (1994)

1. Harris (Freshwater Biology, 32, 143–160, 1994) asserts that while empirical modelling of lake ecosystem properties has yielded general predictions and useful explanations, research into the effects of food‐web interactions has not. I provide an explicit example of a useful, quantitative prediction gleaned from experiments on food‐web effects—the relationship between total phosphorus and the magnitude of response of algal biomass. 2. I further argue that Harris errs in asserting that food‐web effects will be ineffective in eutrophic lakes. This error stems from a misconception about the relevance of ecosystem behaviour in unmanipulated systems for predictions of how an ecosystem will respond to manipulation. I cite empirical evidence demonstrating that, contrary to Harris’s contention, the response of algal biomass to a change in food‐web structure increases as lakes are enriched.