Stable isotopes of S are used in conjunction with dissolved SO2-|4 concentrations to evaluate the utility of ä34S ratios in tracing contributions of bedrock-derived S to SO2-|4 in runoff. Water samples were collected over the annual hydrograph from two tributaries in the West Glacier Lake, Wyoming, catchment. Concentrations of SO2-|4 ranged from 12.6 to 43.0 Ìeq L-1; ä34S ratios ranged from -1.8‰ to +4.9‰ The ä34S value of atmospherically derived SO2-|4 is about +5.6%c.; four samples of pyrite from the bedrock had ä34S ratios that ranged from +0.7 to +4.1‰ Concentrations of SO2-|4 were inversely related to ä34S and discharge. The data for the tributary with the higher SO2-|4 concentrations were reasonably consistent with mixing between atmospheric S and S from a bedrock source with a ä34S ratio of about -4.5‰. The difference from the measured bedrock values presumably indicates that S isotopes in the bedrock pyrite are heterogeneously distributed. The data from the tributary with lower SO2-|4 concentrations did not follow a two-component mixing line. Deviation from a two-component mixing line is most likely caused by preferential elution of SO2-|4 from the snowpack during the early stages of snowmelt, although microbially mediated fractionation of S isotopes in the soil zone also may cause the deviation from the mixing line. Sulfur isotopes are useful in identifying whether or not there is a substantial contribution of bedrock S to runoff, but quantifying that contribution is problematic.