To establish a general model for the removal rate of periphyton by grazers, we identified 27 publications in which removal rates could be estimated from grazer enclosure or exclosure experiments. When all the measurements obtained under different experimental conditions were extracted, these publications provided 107 data points. Multiple regression of these data showed that periphyton removal rate increased significantly with grazer body mass and food availability, and decreased with grazer crowding. Grazer body mass explained 65% of the variation, while crowding and food availability explained 7 and 6% respectively. Except for the significantly lower removal rate of amphibians, neither taxon of the grazer nor algal composition significantly affected removal rate. Experiments in the laboratory and in outdoor channels tended to give higher removal rates than experiments performed in streams or lakes. A comparison with previous allometric equations predicting the ingestion rate of other invertebrate guilds, and with experiments in which periphyton ingestion rate was measured as incorporation of labelled food, indicated that a large portion of periphyton is removed by the activity of the grazer rather than by direct ingestion. These results could be utilized to predict the impact of grazing on periphyton biomass.