CHAPTER XXIV 



THE TECHNIC OF COMPLEMENT-FIXATION REACTIONS 



(Continued) 



Specific Complement Fixation in Bacterial Diseases. As has been 

 stated elsewhere, the first complement-fixation tests were performed by 

 Bordet with bacterial antigens and antiserums (pest and typhoid). 

 Following the application of the principles of complement fixation in the 

 serum diagnosis of syphilis, it was but natural that the possibilities of 

 this method as a general means of diagnosis soon became appreciated, 

 and in a short time numerous infections were studied. 



Probably in no disease has complement fixation proved so constant 

 or so valuable a diagnostic procedure as in syphilis. In this condition 

 the peculiar lipodophilic reagin is largely responsible for the marked 

 fixation of complement, and from our present knowledge on the subject 

 we learn that this phenomenon has practically no analogy in any other 

 disease except frambesia. 



With few exceptions bacterial antigens are likely to yield weaker and 

 more inconstant reactions. This is due to the fact either that our 

 antigen lacks a more available and specific antigenic principle, or that 

 the amount of complement-fixing bodies is small and variable. For 

 these reasons it becomes apparent that the preparation of antigen and 

 delicacy of technic are highly important factors. 



Preparation of Bacterial Antigens. Either the endotoxins or whole 

 bacterial body may constitute the main portion of an antigen. Most 

 recent efforts have aimed thoroughly to disorganize the bacterial cell in 

 order to liberate the endotoxic substances that pass into solution and 

 constitute the antigen. Experience has frequently shown, however, 

 that the protein substances of the bacterial cell itself possess antigenic 

 properties, and accordingly I have generally found that antigens com- 

 posed of cells and the products of cellular activity are usually more 

 satisfactory than those prepared of the endotoxic substances alone. 



As a general rule, bacterial antigens should be polyvalent i. 6., made 

 up of a number of different strains of the same microorganism. Recent 

 researches in bacteriology tend to show that different strains of the 



498 



