300 BACTERIOLOGY 



prove this "fixation" of complement in the first mixture: to 

 this end, after the mixtures had stood for a time, an incom- 

 plete hemolytic system was added to each mixture that is, 

 an amount of normal washed blood corpuscles and a portion 

 of inactivated immune serum otherwise hemolytic for those 

 corpuscles, was added. Before this addition, obviously, 

 no hemolysis could occur, because the complement of the 

 hemolytic serum had been destroyed by the heat used for 

 inactivation. But after the addition hemolysis did occur in 

 one tube but not in the other. It is plain that complement 

 necessary to the phenomenon of hemolysis must have been 

 available in one of the tubes. If one recalls that in the second 



FIG. 61 FIG. 62 



= Blood corpuscle. Cj = Blood corpuscle. 



Lv/J = Hemolytic amboceptor. 

 = Hemolytic amboceptor. 



f^M) = Complement. 



No hemolysis. No complement Hemolysis. Free complement of 



available; all fixed, as in a', original mixture now bound by 



hemolytic amboceptor. 



mixture no immune bodies or amboceptors specifically 

 related to the antigen were present it is clear that the com- 

 plement could not have been bound or fixed. It must have 

 remained free in the serum, available for complementing 

 the action of the hemolytic amboceptors and thereby hemo- 

 lyzing or destroying the normal blood corpuscles added, as 

 shown by the laking of such corpuscles in the tube. This, 

 in short, is what occurred. See Figs. 61 and 62. 



For this particular test, Bordet and Gengou used plague 

 antigen (plague bacilli); plague amboceptors (present in 

 blood of animal immunized from plague); complement (free 

 in normal mammalian blood); normal serum (containing no 



