1. In these studies several phases of the predpitm reactions were investigated by the use of purified proteins as antigens. These preparations were edestin from hemp-seed and crystalline ovalbumin from fresh eggs. The ovalbumin, isolated by the method of Hopkins and Pinkus,.was apparently as pure as is obtainable by chemical means. This albumin, however, produced moderately severe anaphylactic reactions in animals sensitized with ovoglobulin. Anaphylactic tests of the individuality of a protein cannot be any longer regarded as the criterion of the purity of the substance as an antigen.
Wells and Osborne have shown that proteins of considerable chemical difference may have a common antigenic group which causes mutual anaphylactic reactions in animals sensitized to these proteins. In particular, as egg globulin is a mixture of proteins, one of which is undoubtedly egg albumin, anaphylaxis produced by injections of albumin into animals sensitized to the so called globulin offers no evidence for or against the purity of the albumin. The character of the curves shown in Text-fig. 1 confirms the assumption, based upon chemical data, that crystalline egg albumin is a single protein.
2. With edestin and crystalline egg albumin as antigens phases in the precipitin reaction were found in which these substances and their specific precipitins could be demonstrated to be coexistent but ununited in the same serum.
3. When edestin or crystalline egg albumin is injected into a rabbit immunized thereto, the antigen may be found in the circulating blood during 48 hours after its injection, while at the same time the animal maintains a high titer of free precipitin in its blood.
4. When the pure protein antigen is mixed in proper proportions with the serum of a specifically immunized rabbit and the resulting precipitate removed by centrifugation, the supernatant fluid contains both antigen and antibody.
5. The serum drawn from a rabbit during the period in which free antigen and antibody are coexistent in the circulation undergoes slow spontaneous precipitation when kept in sterile tubes in the ice box. The curve of this reaction is reproduced as Text-fig. 1. The relationships of the parabola indicate that the interaction of antigen and antibody takes place according to a definite law. When sufficient quantitative data are obtained to allow an analysis of this curve, the formulas for this reaction will undoubtedly throw light upon the chemical or physical nature of the process.
6. The protective action of the solution of egg albumin as a third colloid preventing precipitation in a reaction between human serum and its antibody was readily demonstrated. This observation and the constancy of the long prozone in precipitin test with egg albumin are in accord with the protective action of ovalbumin upon colloidal gold.