The relative efficacy of thiol-containing mercurial scavengers was assayed by using cellular suspensions of erythrocytes or isolated hepatocytes. The blood cells incubated in a buffer (pH 7.4) containing 1 mM glucose (10% hematocrit) were exposed to 5 microM methyl mercuric chloride. In the absence of extracellular thiols the red blood cells took up more than 90% of methyl mercury from the surrounding medium during 5--10 min. This uptake was almost completely inhibited by dimercaptosuccinic acid (DMSA) (1 mM) and the same chelant could rapidly remove 80% of the mercury from 'pre-loaded' erythrocytes. Hepatocytes prepared according to the method of Seglen [11] in a suspension of 10(6) cells/ml in a buffer containing 5 mM glucose and 5 mg/ml of bovine serum albumin were also exposed to methyl mercuric chloride (4 microM). Almost 50% of the mercurial was taken up by the cells slowly during the incubation period of 240 min. DMSA (1 mM) almost completely blocked the methyl mercury binding by the hepatocytes. 2-Mercaptopropionylglycin (Thiola) or mercaptosuccinic acid (MSA) was almost as effective mercurial scavengers as DMSA in hepatocytes and in red blood cells. Diethyldithiocarbamate (DDC) and dimercaptopropanol (BAL) were considerably less effective than DMSA to inhibit the mercurial binding to hepatocytes. Experiments in vivo have shown that DMSA is a better mercurial chelator than Thiola or MSA, whereas DDC and BAL may both be considered to be inapplicable in methyl mercury poisonings. Our cellular assay provides preliminary information of the efficiency of chelating thiols and may serve as a useful first approximation when planning further experiments.