Studies with goat and rabbit anti-mouse antibody, as models of human anti-mouse antibody (HAMA), have shown that several of the commonly used two -site immunoassays (e.g., for CA-125, carcinoembryonic antigen, choriogonadotropin, lutropin, hepatitis B surface antigen, thyrotropin) are susceptible to interference by this type of bridging heterophile antibody. In most cases the interference can be blocked by incubation with mouse IgG. We studied HAMA interference in an assay of hepatitis B surface antigen by using HAMA-positive sera from a transplant patient given OKT3 and from a cancer patient given CYT-103 (modified antibody B72.3). The HAMA interference attributable to the OKT3 could be blocked by incubation with mouse IgG at room temperature. In contrast, the interference caused by the B72.3-induced HAMA could be blocked by prolonged incubation with high concentrations of the B72.3 antibody at 4 degrees C. A limited survey of 50 hospital patients selected without conscious bias revealed two HAMA-positive patients, only one of whom was known to have been exposed to mouse immunoglobulin. HAMA interferences are currently a minor problem in routine laboratory medicine, but the increasing use of diagnostic and therapeutic products involving mouse-origin monoclonal antibodies will make the detection and elimination of HAMA interferences an important part of laboratory practice in the future.