290 INFECTION AND IMMUNITY. 



tive, that is, there is no evidence of .syphilis. If 

 hemolysis is absent as in the positive control tube 

 the reaction is characteristic of syphilis. 



Various modifications of this technic have been 

 used by different experimenters with good results. 

 Noguchi, for instance, has used human blood sus- 

 pensions with their amboceptor instead of sheep's 

 blood, the purpose being to avoid complications 

 due to hemolysins for sheep's corpuscles found 

 occasionally in human serum. It has been found 

 that a larger number of positive reactions in 

 syphilis can be obtained by using non-heated 

 serum of the suspected patient. It is also found, 

 however, that a larger number of normal inactive 

 serums react as positives; the reaction is therefore 

 not so specific as when the heated serum is used. 

 Synthetical antigens composed of mixtures of 

 lipoid bodies of known composition, have been used 

 with varying degrees of success. 



value of the The Wassermann reaction has proved to be of 

 Wa Slae l t?o a n t . reat value in tne diagnosis of syphilis. The per- 

 centage of positive reactions found in the various 

 stages of syphilis varies with the technic of differ- 

 ent observers. The following table is made from 

 a collection of percentages by Pearce: 



Highest Lowest 



Stage of Disease per cent, per cent. 



Primary syphilis 92.8 64.4 



Secondary syphilis 100 71 



Tertiary syphilis 100 63 



Early latent syphilis 76 51 



Late latent syphilis 79 46 



Hereditary syphilis 100 86 



The percentage of cases in which syphilis could 

 be excluded so far as possible from history and 



