BACTERIAL POISONS. 191 



bacteria fail to clump in the blood of patients suffering from a 

 given disease. Again, in some cases the serum in a certain 

 disease will clump bacteria that are not concerned in the 

 production of the disease. Even normal serum will sometimes 

 clump bacteria. The serum from a given disease is said to be 

 homologous with the bacteria causing the disease and heter- 

 ologous from other bacteria, and the bacteria are said to 

 clump or not to clump with homologous sera as the case may 

 be. Bacteria are also homologous or heterologous in the same 

 sense toward sera. 



Park and Collins,* in summing up their experience in the 

 study of agglutinins among other things, state that the injec- 

 tion of animals with bacteria causes the production of agglu- 

 tinins which are specific for the organism used for injection, 

 but in addition to this it also causes the production of agglu- 

 tinins which agglutinate bacteria other than those injected; not 

 only closely allied bacteria, but those which differ widely in their 

 biochemical as well as in their specific agglutinating characteris- 

 tics. The injection therefore forms specific, group and heterolo- 

 gous agglutinins. Of these there is at first more specific agglu- 

 tinin than of the others. Further injection may cause a more 

 or less rapid diminution of the specific agglutinin along with 

 a less rapid diminution of the others, so that the relative 

 amounts of the different agglutinins may vary. The heter- 

 ologous agglutinins may be absorbed by injecting the immune 

 animal with various bacteria other than the one used for the 

 immunization so as to free the serum from all but the specific 

 agglutinin so that it will clump only the bacteria which were 

 first used for the immunization. Group agglutinins may also 

 be absorbed by injecting the immune animal with bacteria 

 nearly allied to that used for immunization. Thus the typhoid 

 bacillus will absorb many group glutinins for the colon bacil- 

 lus. The injection of bouillon into rabbits will frequently 



Journ. Med. Research. XII., pp. 491-507. 1904. 



