BACTERIAL COMPLEMENT FIXATION TECHNIQUE 415 



and bacterial agglutinins, respectively. It is realized that some of 

 the variation may be due to variations in the time interval for 

 maximum antibody production, but it is certain that some animals 

 are entirely refractory as judged by our present methods of in- 

 vestigation. 



In the commercial j^roduction of diphtheria antiloxiji, hoi-scs 

 are given repeated injections oC toxin for the most pai-t in the 

 form of toxin-antitoxin mixtures or of anatoxin, which is toxin 

 detoxified with formaldehyde. It has been the general experience 

 that some horses yield little or no antitoxin while others produce 

 high titered antitoxic sera with all degrees of variation between 

 these two extremes. 



Antibody Variation During Infection. — Likewise, the physio- 

 logical production of antibodies resulting from infection is variable 

 even when the infectious agent is a good antigen. This is quite 

 definiteh' illustrated by a study of the agglutinin response in nor- 

 mal rabbits infected with P. tularensis producing a disease com- 

 mon to rabbits. In a series of 11 normal rabbits, showing zero 

 agglutinins before inoculation, and all surviving for almost the 

 same length of time. Downs observed that one developed a titer 

 of 1 :1,200, three of 1 :320, one of 1 :160, two of 1 :40, two of 1 :20 

 and two continued with a zero titer until death. In the case of 

 typhoid fever, the literature indicates that the Widal reaction is 

 positive at some time during the disease in 80 to 92 per cent of the 

 cases. Gay reported 100 per cent positive Ijy the fifth week in a 

 series that he studied. It should be remembered that E. typhosa 

 is an unusually good antigen represented by practically one type 

 in contrast to the numeroiLS groups or types of pneumococci, 

 gonocoeci, and meningococci found associated with their respective 

 infections. 



Summary of Recommendations. — In this chapter some of the 

 factors influencing the results of bacterial complement fixation and 

 the underlying philosophy of hemolysin, complement and antigen 

 preparation and titration have been presented by discussion and 

 protocols. In conclusion it would seem advisable to summarize 

 most of the specific recommendations that have been presented rela- 

 tive to a standard procedure for complement fixation. 



It will be observed that these conform very closely to Kolmer's 

 recommendations : 



