THE TECHNIQUE OF SERUM REACTIONS 265 



tion with normal serum, is the one to be employed in subsequent re- 

 actions. Before actual use, it is convenient to make a dilution of antigen 

 in salt solution in such a way that 1 c.c. shall contain the amount re- 

 quired. Thus if 0.05 c.c. is wanted, mix 0.5 c.c. with 9.5 c.c. salt solu- 

 tion. Then 1 c.c. of this can be added to each tube in the test. 



II. The Hemolytic Serum. — ^The hemolytic amboceptor, for the 

 reaction, is obtained by injecting into rabbits the washed red blood 

 corpuscles of a sheep. A 5 per cent emulsion of the corpuscles is made 

 and of this 5 c.c, 10 c.c, 15 c.c, etc., are injected at intervals of five 

 or six days. Three or four graded injections of this kind are usually 

 sufficient to furnish a serum of adequate hemolytic power. The injec- 

 tions may be made intraperitoneally or intravenously. About nine or 

 ten days after the last injection of corpuscles, the rabbit is bled from the 

 carotid artery and the serum obtained by pipetting it from the clot. 



It is best to have a hemolytic serum of high potency in order that the 

 quantities used for the reaction may be as small as possible. This is 

 desirable because of the fact that the serum may contain small amounts 

 of precipitins for sheep's serum, due to insufficient washing of the cor- 

 puscles employed in the immunization. If such precipitins should be 

 present in any quantity in the serum used for the reaction, precipitates 

 might be formed, and these, as we know, have a tendency to carry down 

 complement from a mixture. 



WhUe the quantitative relations of the complement and antigen in 

 the Wassermann reaction are important, they are vastly more so in the 

 case of the hemolytic amboceptor. For the actual reaction most 

 observers make use of two hemolytic units. A hemolytic unit is the 

 quantity of inactivated immune serum which, in the presence of com- 

 plement, suffices to cause complete hemolysis in 1 c.c. of a 5 per cent 

 emulsion of washed blood corpuscles. Noguchi^ has pointed out very 

 clearly the dangers of not delicately adjusting the quantity of ambo- 

 ceptor used in the reaction. In a recent communication upon the 

 subject, he has called attention to the experiments of Morgenroth and 

 Sachs ^ who have shown that the relationship between complement and 

 amboceptor necessary for hemolytic reactions is one of inverse propor- 

 tions. To state it more clearly, in their own words, "in the presence of 

 larger quantities of amboceptor, smaller quantities of complement suf- 

 fice," and vice versa. Noguchi, in his work, has found that, while, in the 



1 Noguchi, Proc. Soc. for Exper. Biol, and Med., VI., 3, 1909. 



'Morgenroth und Sachs, in Ehrlich's " Gesammelte Arbeiten," etc., Berlin, 1904, 



