Sulfur isotope analysis (δ34S) of well-preserved carbonates spanning an ~10 Ma interval of the terminal Proterozoic Nama Group reveals that disseminated pyrite is consistently enriched in 34S relative to coeval seawater sulfate as preserved in carbonate-associated sulfate (CAS). This observation is not consistent with the current paradigm for interpreting the geologic record of sulfur isotopes, which assumes that pyrite δ34S (δ34Spyr) will be equal to or less than co-occurring CAS δ34S (δ34SCAS) due to the kinetic isotope effect of bacterial sulfate reduction (BSR) that favors the lighter isotope of sulfur (32S) during sulfur-oxygen bond breakage. Although the precise mechanism of pyrite sulfur isotope enrichment is debatable, our combined observations of extremely 34S-enriched pyrite, low bulk-rock concentrations of sulfur from CAS and pyrite, and high-frequency fluctuations in δ34SCAS and δ34Spyr throughout the Nama Group carbonates point to very low concentrations of sulfate in portions of the terminal Proterozoic ocean. The additional occurrence of 34S-enriched pyrite in contemporaneous terminal Proterozoic sections from Poland and Canada reveal that low seawater sulfate may have been widespread in the oceans at this time. However, the absence of such extremely 34S-enriched pyrite from well-preserved, coeval carbonate sections in Oman suggests that such conditions were not globally uniform. Low, geographically varied concentrations of marine sulfate in terminal Proterozoic time are consistent with elevated, geographically varied concentrations of reactive, nonpyritized iron in marine shales recently reported for this interval, suggesting that high dissolved Fe(II) and low O2 persisted in a range of marine facies as late as 543 Ma—tens of millions of years after the origin of animals, yet prior to their major diversification in Early Cambrian time.