688 



INFECTION BY THE BLOOD PROTOZOA 



An important development of this work on trypanolysis which has many theo- 

 retical ramifications and which may explain some features of relapse phenomena is 

 that there must be a definite relationship between the amount of immune serum and 

 the occurrence of trypanolysis. Thus, paradoxically, if too much immune serum is 

 given, there is no trypanolysis. This is obviously similar to the so-called "zone 

 phenomenon" or the phenomenon of Neisser and Wechsberg (1901) who, during the 

 estimation of the bactericidal property of immune serum in vitro, made the observa- 



TABLE I 



Experiment with Immune Serum from Infected Rabbit (from Taliaferro and Johnson) 

 Trj^anosomes Injected Intraperitoneally into Each Mouse at Beginning of Experiment* 



* Plus sign indicates presence of trypanosomes in the blood; minus sign indicates their absence. No attempt is made 

 to indicate the increase in parasites because once they appear in the blood they increase uniformly until death or until an 

 artificial crisis is produced. During the relapse, after the artificial crisis, the same holds true. 



tion that in the presence of constant amounts of complement and antigen (Vibrio 

 metchnikovi) moderate amounts of antiserum were completely bactericidal, whereas 

 very large amounts of antiserum were no more bactericidal than small amounts. The 

 phenomenon is of common occurrence in in vitro serological titrations and in various 

 in vivo tests of the protective properties of antibacterial sera. 



Furthermore, we found that there are recurring zones of action and inaction of 

 the lytic serum as shown in the experiment in Table I. Here fifteen mice were all 

 infected at the same time with the passage strain. When they showed from ten to 

 twenty trypanosomes per standard field (see third day of infection in table), one was 

 left untreated as a control, while the other fourteen were given various amounts of 



