In developing sediment quality criteria (SQC) for metals, it is essential that bioavailability be a prime consideration. Different studies have shown that while dry weight metal concentrations in sediments are not predictive of bioavailability, metal concentrations in interstitial (pore) water are correlated with observed biological effects. A key partitioning phase controlling cationic metal activity and toxicity in the sediment-interstitial water system is acid-volatile sulfide (AVS). Acid-volatile sulfide binds, on a mole-to-mole basis, a number of cationic metals of environmental concern (cadmium, copper, nickel, lead, zinc) forming insoluble sulfide complexes with minimal biological availability. Short-term (10-d) laboratory studies with a variety of marine and freshwater benthic organisms have demonstrated that when AVS concentrations in spiked or field-collected sediments exceed those of metals simultaneously extracted with the AVS, interstitial water metal concentrations remain below those predicted to cause effects, and toxicity does not occur. Similar observations have been made in life-cycle laboratory toxicity tests with amphipods and chironomids in marine and freshwater sediments spiked with cadmium and zinc, respectively. In addition, field colonization experiments, varying in length from several months to more than 1 year, with cadmium- or zinc-spiked freshwater and marine sediments, have demonstrated a lack of biological effects when there is sufficient AVS to limit interstitial water metal concentrations. These studies on metal bioavailability and toxicity in sediments serve as the basis for proposed SQC for the metals cadmium, copper, nickel, lead, and zinc. Specifically, four approaches for deriving criteria are described: (a) comparison of molar AVS concentrations to the summed molar concentration of the five metals simultaneously extracted with the AVS; (b) measurement of interstitial water metal concentrations and calculation of summed interstitial water criteria toxic units (IWCTU) for the five metals, based upon final chronic values from water quality criteria documents; (c) calculation of summed IWCTU based upon sediment AVS concentrations and metal-specific partitioning of the metals to organic carbon; and (d) calculation of summed IWCTU based upon partitioning of the metals to a minimum binding phase sorbent (chromatographic sand). For a number of reasons, SQC derived using these approaches generally should be considered “no effect” values, i.e., with these techniques it is possible to predict when sediment metals will not be toxic, but not necessarily when metal toxicity will be manifested. Currently, approaches (a) and (b) are the most useful in terms of predicting metal bioavailability and deriving SQC. Further research is required, however, to fully implement approaches (c) and (d). Additional research also is required to thoroughly understand processes controlling bioaccumulation of metals from sediments by benthic organisms, as well as accumulation of metals by pelagic species that ingest metal-contaminated benthos.