The determination of acid-volatile sulfide (AVS) and simultaneously extracted metals (SEMs) in sediment by treatment with dilute HCl shows promise as a tool for predicting the potential for metal toxicity to sediment-dwelling organisms. Effective quality control measures must be developed if this method is to become a reliable procedure and to ensure comparability of data. However, establishing quality control measures that assess procedural errors for an operationally defined method can be problematic. For example, preextraction spikes added for assessing the accuracy of AVS and SEMs may be poorly recovered due to adsorption or reaction with sediment constituents. For a variety of sediment types, we found preextraction spikes of sulfide, mercury, and copper to be prone to variable recoveries for the AVS/SEM procedure; recoveries averaged 76.3% (SD, 20.9) for sulfide, 61.9% (39.6) for Hg, and 90.1% (12.7) for Cu. The average recovery was near 100% for preextraction spikes of sediments for Cd, Ni, Pb, and Zn, and the recoveries of preextraction blank spikes for all analytes were consistently 95 to 105%. Binding of Cu or Hg with sulfides is sufficiently strong that 1 N hydrochloric acid will not necessarily keep the spiked metal in the dissolved state. This does not mean that the SEM procedure is invalid for these metals, only that the quality control of procedural error is difficult to assess. However, Hg will generally not be detected when measured as an SEM because of its tendency to adsorb onto sulfide minerals even at extremely low pH. Some reference sediments may be useful for assessing consistency of AVS determinations; we measured 5.97 ± 0.65 μmol/g in National Institute of Standards and Technology (NIST) 1645 and 1.34 ± 0.14 μmol/g in NIST 2704 for repeated determinations conducted over the past 3 years. Apparently, some sediments may contain an oxidation-resistant sulfide component that can release low to moderate AVS when treated with dilute HCl.