In China, coal combustion is the most important source of atmospheric sulfur pollution. Moss sulfur isotopic signatures have been believed to hold source-specific information that can serve as a fingerprint to identify atmospheric sulfur sources. In cities where only local coals were combusted, we observed a good correspondence of average sulfur isotope ratios in urban mosses (Haplocladium microphyllum) to the values of local coals (δ(coals) = 1.455δ(mosses)- 3.945, R(2) = 0.975, p = 0.01). But if different types of coals were combusted, we did not know whether moss sulfur isotope ratios can indicate mixed coals. To confirm this, using a mixing model we estimated the ratios of imported coal to local coals at cities where both coals were used. We found that the estimated ratios at large cities (>1 million people) where both coals were used were similar to the reported ratios in their respective provinces. For small cities (<0.5 million people) in Jiangxi Province and other provinces, the estimated ratios were higher than the reported ratios because the relatively cheaper local coals were less used in all the small cities except in cities where local coal deposits were found nearby. The comparison results showed that moss sulfur isotope is a useful tool for indicating coal-derived sulfur even in cities where mixed coals were combusted.