Stones of different rock types often accrue different amounts of periphytic algae. Although algal biomass may be positively related to stone roughness, the confounding role of rock chemistry is unclear. This independent effect of rock chemistry on benthic algae was tested using the nutrient-diffusing technique, by incorporating powdered stone, rather than nutrients, into the agar matrix. Rocks tested were sandstone, obsidian, schist, greywacke, pumice, gypsum, limestone, serpentine, and phosphorite. Petri-dishes containing powdered rock and agar, and covered with a permeable cellulose filter, were incubated in eight pools in a granitic stream. Algal biomass did not differ among any of the nine rock types and plain agar control, whereas biomass differed among the concurrently placed nutrient diffusing substrates (the stream was phosphorus-limited). Algal composition was more related to an upstream-downstream gradient (for filamentous algae) and pool-specific effects (deposition of fine sediment for diatoms) than rock chemistry. This minimal effect of rock chemistry may be caused by the low dissolution rate of stones.