1. The globulins of both normal and diphtheria antitoxic serum exhibit chemically toward reagents the same reactions, being precipitated by magnesium sulphate and split up into fractions in precisely the same way.
2. All of the diphtheric antitoxic power of both normal and immunized serum is always carried by the globulin and its fractional precipitates.
3. During the fractional precipitation of the serum globulin of horses immunized from diphtheria toxin and horses not immunized from diphtheria toxin, some of the globulin is lost, likewise at the same time some of the antitoxic power of the globulin of the immunized serum is lost.
4. These reactions, considered in connection with the fact that different observers as well as we ourselves have found diphtheric antitoxic power in normal horse's serum and that this antitoxin separates with the globulin, strongly incline us to consider "diphtheria antitoxin" a form of globulin.
5. The reactions of globulin, previously separated from the serum by magnesium sulphate, with sodium chloride lead one to think that there is a formation of globulin salts.
6. Since serum albumin in a magnesium sulphate solution gives fractional precipitates at definite temperatures, it seems not improbable that the albumin is precipitated in the form of albumin salts.