Understanding the role of soil microbial communities in coupled carbon and nitrogen cycles has become an area of great interest as we strive to understand how global change will influence ecosystem function. In this endeavor, microbially explicit decomposition models are being adopted because they include microbial stoichiometry- and biomass-mediated mechanisms that may be important in shaping ecosystem response to environmental change. Yet there has been a dearth of empirical tests to verify the predictions of these models and hence identify potential improvements. We measured the response of soil microbes to multiple rates of carbon and nitrogen amendment in experimental microcosms, and used the respiration and nitrogen mineralization responses to assess a well-established, single-pool, microbial decomposition model. The model generally predicted the empirical trends in carbon and nitrogen fluxes, but failed to predict the empirical trends in microbial biomass. Further examination of this discontinuity indicated that the model successfully predicted carbon and nitrogen cycling because stoichiometry overrode microbial biomass as a regulator of cycling rates. Stoichiometric control meant that the addition of carbon generally increased respiration and decreased nitrogen mineralization, whereas nitrogen had the opposite effects. Biomass only assumed importance as a control on cycling rates when stoichiometric ratios of resource inputs were a close match to those of the microbial biomass. Our work highlights the need to advance our understanding of the stoichiometric demands of microbial biomass in order to better understand biogeochemical cycles in the face of changing organic- and inorganic-matter inputs to terrestrial ecosystems.