140 



Scientific Proceedings (75). 



which has daily been balanced by simultaneous alexin and he- 

 molysin titrations. Pooled serum from six or more guinea pigs 

 has been preserved by salting for alexin, a method which has 

 many advantages. 



Serum cross-titrations with standardized antigens have de- 

 veloped the fact that different strains of typhoid bacilli fall into 

 different groups, somewhat analogous to the groupings of pneu- 

 mococci. This is evidence which must be seriously considered in 

 the explanation of the causes of failure in typhoid vaccination, 

 since it is not unlikely that subjects may have been infected with 

 strains dissimilar to those used for prophylactic inoculation. The 

 strains tentatively placed in Group I cross-fix with all antigens; 

 those in Group II cross-fix with each other but not with Group I 

 strains. Group I-A strains give irregular results. There is no 

 apparent relation between the virulence of the organisms as in- 

 dicated by the severity of the disease which they caused, or between 

 their toxicity as indicated by the reactions produced in immunizing 

 rabbits, and this grouping. All of the Group II strains, however, 

 have been under artificial cultivation for a number of years. 

 Number 13 is the Rawlins strain, so extensively used in prophy- 

 lactic immunization. It seems to possess a lesser antigenic com- 

 plexity than does any organism in the other groups. On this 

 evidence it would seem to be theoretically of less value as an 

 immunizing strain. 



The results of absorption tests, while somewhat less consistent, 

 are confirmatory of the findings obtained with the alexin fixation 

 reaction. Group I and Group II serum absorbed respectively 

 with Group I and Group II strains give usually negative results 

 with all organisms. Absorption of Group I sera with Group II 

 strains results in the removal of all agglutinins for Group II while 

 agglutinins for Group I still remain. 



It is considered that the evidence of antigenic differences thus 

 far discovered among typhoid strains is sufficiently valid to warrant 

 the presentation of these data, and sufficiently encouraging to 

 justify similar more extensive work, especially with regard to the 

 comparative protection which is aroused by strains of different 

 character. Although the principle of polyvalency has been used 

 empirically in the past, it would seem advisable now on more 



