Many soils of the McMurdo Dry Valleys (MDV), Antarctica, being old, hyperarid, and frigid, have accumulated abundant atmospheric salts over the last several million years. This salt repository offers an opportunity to study atmospheric chemistry (past and present), the origin and transport of ions in soils, the weathering activity of soils, and postdepositional soil-leaching processes within this unique environment. In particular, soil sulfate in the MDV is known to have multiple origins, but the precise proportions of different sulfate components remain elusive. Here we test a hypothesis that soil sulfate in the MDV is a predictable mixture of three major components: sea-salt sulfate, non-sea-salt sulfate, and background sulfate (derived from weathering and volcanic sources) that to a large extent, varies as a function of elevation and distance from the coast. By measuring sulfate's three stable isotope parameters, i.e., Δ17O, δ18O, and δ34S, plus estimating independently the end-member stable isotope parameters for the three components on the basis of published reports and our measured data, we solve explicitly the mixing proportion for each component using a set of three simultaneous linear equations. The results from four spatially representative soil profiles show that the sum of such calculated proportions for the three components is very close to unity in most samples, suggesting that a simple three-component mixing model is a good representation of the soil sulfate budget. While significant uncertainties still exist in the isotope compositions of different sulfate end-members, these data provide important initial constraints on sulfate diversity as well as its spatial and vertical distribution in the MDV.