This method is usually spoken of as the “complement-fixation” or the “complement-deviation test” (“Wassermann test” in syphilis) and is applicable in a great variety of microbial diseases, but it is of practical importance in those diseases only where other methods are uncertain--syphilis in man, concealed glanders in horses, contagious abortion in cattle, etc. A better name would be the “Unknown Amboceptor Test” since it is the amboceptor that is searched for in the test by making use of its power to “fix” complement. The principle is the same in all cases. The method depends, as indicated above, on the ability of complement to combine with at least two amboceptor-antigen systems, and on the further fact that if the combination with one amboceptor-antigen system is once formed, it does not dissociate so as to liberate the complement for union with the second amboceptor-antigen system. If an animal is infected with a microörganism and a part of its defensive action consists in destroying the organisms in its blood or lymph, then it follows from the above discussion of cytolysins that there will be developed in the blood of the animal amboceptor specific for the organism in question. If the presence of this _specific amboceptor_ can be detected, the conclusion is warranted that the organism for which it is specific is the cause of the disease. Consequently what is sought in the “complement-fixation test” is a _specific amboceptor_. In carrying out the test, blood serum from the suspected animal is collected, heated at 56° for half an hour to destroy any complement it contains and mixed in definite proportions with the specific antigen and with complement. The antigen is an extract of a diseased organ (syphilitic fetal liver, in syphilis), a suspension of the known bacteria, or an extract of these bacteria. Complement is usually derived from a guinea-pig, since the serum of this animal is higher in complement than that of most animals. The blood of the gray rat contains practically as much. If the specific amboceptor is present, that is, if the animal is infected with the suspected disease, the complement will unite with the antigen-amboceptor system and be “fixed,” that is, be no longer capable of uniting with any other amboceptor-antigen system. No chemical or physical means of telling whether this union has occurred or not, except as given below, has been discovered as yet, though doubtless will be by physico-chemical tests, nor can the combination be seen. Hence an “indicator,” as is so frequently used in chemistry, is put into the mixture of antigen-amboceptor-complement after it has been allowed to stand in the incubator for one-half to one hour to permit the union to become complete. The “indicator” used is a mixture of sheep’s corpuscles and the heated (“inactivated”) blood serum of a rabbit which has been injected with sheep’s blood corpuscles and therefore contains a _hemolytic amboceptor specific_ for the corpuscles