Dissimilatory nitrate reduction to ammonium (DNRA) involves the conversion of nitrate to ammonium, in contrast to the conversion of nitrate to N₂, as in denitrification.  The resultant ammonium is not permanently removed, but retained within the ecosystem.  Two forms of DNRA are known to occur: fermentative and sulfur-oxidizing DNRA.  Some have hypothesized that carbon controls fermentative DNRA; however, others have hypothesized sulfide controls DNRA by inhibiting key enzymes.  We tested the relative importance of carbon vs. sulfide in regulating DNRA using a microbial assay.  This method involves adding nitrate along with a gradient of carbon and sulfide to anoxic sediments from a high ambient sulfide site and a low sulfide site.  The addition of carbon and sulfide interacted with site differences to affect nitrate removal (p<0.05).  Nitrate was removed more rapidly and N₂ production was 4x greater in the low sulfide site than the high sulfide site (p=0.00).  Ammonium production accounted for ~30% of the nitrate loss in the high sulfide site and 0% in the low sulfide site.  Preliminary analysis indicates that sulfide may control the production of both N₂ and ammonium.  The potential importance of DNRA has profound implications for our understanding of N cycling in aquatic ecosystems.