Summary: The general findings of Longcope and Rackemann, as to the presence of circulating antibodies following serum disease and their absence following treatment with antisera without serum disease, have been confirmed by this study. In addition, we have shown that antibody content is conditioned by the intensity of the disease and by the interval after the onset of the disease at which the bleeding is made; that anaphylactic antibody is more regularly present than has been previously indicated; that the positive Prausnitz-Küstner reaction may be demonstrated in a large proportion of cases; and that the passive sensitization of the guinea pig ear-skin could be regularly induced. A definite relationship between the antibody responsible for the Prausnitz-Küstner reaction and the basic mechanism of serum disease is not clear; our results indicate no more than that this antibody occurs as one phase of a general stimulation of the various antibodies, initiated by a cellular reaction to the presence of a foreign serum; that it differs from an atopic reagin, in its experimental manifestations, only in that it persists in the serum of the patient but for a short time; and that the finding of this antibody in serum disease sera may be regarded as an instance of reagin production to a typical atopen in non-atopic individuals.