Three heterophile antibody tests and a test specific for IgM antibody to Epstein-Barr virus were evaluated during prospective studies of infectious mononucleosis. Specificity was judged by the frequency of false-positive results in sera of known qualities taken before illness; except for two patients bled during early, unrecognized illnes,, titers of greater than or equal to 1:40 were detected in 12% by the absorbed sheep red cell test, in 6.7% by the absorbed horse red cell test, and in none by the beef cell hemolysin test. None had IgM antibody specific for Epstein-Barr virus in sera obtained before illness. In addition, no rises in titer of heterophile antibody were detected by the horse cell test in 38 patients with proved rubella and/or influenza infection. In terms of sensitivity (indicated by the percentage of cases with diagnostic titers during infectious mononucleosis), 97% were positive by the Epstein-Barr virus IgM test, 96% by the horse cell agglutination test, 85% by the beef hemolysin test, and 81% by the sheep cell agglutination test. Persistence of antibody was judged by serial bleedings up to three years after illness; titers of heterophile antibody by the sheep agglutination and beef hemolysin tests as well as titers of IgM antibody to Epstein-Barr virus returned to normal in two to three months, whereas the horse cell heterophile test remained positive for a year or more in 75%. Inapparent and mild infections with Epstein-Barr virus resulted in the production of horse cell heterophile antibody in 48.4% of 122 subjects.