BLUMBERG and his colleagues have published data on the relation of Australia antigen, Au(1), to hepatitis1,2. While existing antisera, either of human or rabbit sources3, frequently can be used to detect this antigen in patients with hepatitis by either immunodiffusion4 or complement fixation5 it is probable that the antigen is undetectable in some patients. One possible reason for this may be the neutralization of antigen by antibody. Sera may manifest the presence of antigen at one time and antibody at another time2. In such a case, a precipitin band forms between sera from early and later bleedings from the same patient.