Twenty reference sera with previously reported titers plus 10 normal seropositive and 10 seronegative sera were used to determine which assay for the detection of antibodies to Epstein-Barr virus should be utilized in our laboratory. By three commercially available test systems and by slides prepared in our laboratory, all techniques detected the same number of samples positive for IgG antibodies to viral capsid antigen (VCA) and Epstein-Barr virus-determined nuclear antigen (EBNA), but the geometric mean titers (GMTs) varied. Major discrepancies arose in the sensitivity of detection of early antigen-diffuse (EA-D) and -diffuse and restricted (EA-DR). Slides prepared in our laboratory gave the best results, detecting 12 of 12 (100%). Slides prepared according to Gull and Hillcrest detected 11 of 12 (92%), and those prepared according to Zeus detected 9 of 12 (75%). From these data, we conclude that all of these methods for detection of VCA and EBNA are adequate. However, in determining antibody levels to EA-D and EA-DR, great care should be taken as to the source of the slide and specificity and sensitivity.