314 INFECTION AND IMMUNITY 



other in the natural classification. An explanation of reactions to a 

 given infection which applies in the case of one species is not, therefore, 

 obviously applicable in the case of another species. This is true not only 

 of the mechanism of protection as it takes place in the serum of dif- 

 ferent animals and in their plasma, but also of phagocytosis and phagocy- 

 tic digestion and the factors which contribute to the perfection of these 

 processes. The constant stumbling-block in the way of a correct in- 

 terpretation of processes going On in the animal body is our inability, as 

 we heve seen, to argue from serum phenomena to phenomena occurring 

 in the plasma. A failure to keep this in mind, although it is fully recog- 

 nized, has undoubtedly led to many hasty conclusions, particularly 

 connected with the theory of lytic immunity. This may be illustrated 

 by a well-known example : Fresh rabbit serum is actively germicidal for 

 anthrax bacilli, dog serum is not; yet rabbits are extremely sensitive 

 to a true anthrax infection, while dogs are very resistant. Experiment 

 has shown that there are tytic amboceptors in the sera of both these 

 animals, but that the dog's serum does not contain the complement 

 necessary for their action on the bacilli; the complement presumably 

 has remained in the body cells, whereas in the case of the rabbit it 

 has possibly been liberated from the leucocytes during clotting. The 

 reason the dog is insusceptible is, then, not because of a more active 

 plasma destruction of the invading ' anthrax germs, but probably 

 because of a more perfect adjustment of the cellular mechanism to the 

 infection, although if we simply followed the theory of the bactericidal 

 action of serum and plasma as being coextensive, and the active pro- 

 tective mechanism, the rabbit should have been protected, while the dog 

 should have succumbed. The difference here probably depends upon 

 the possession of all requisites for the perfect performance of phagocy- 

 tosis, and the complete digestion of the bacteria by the phagocytes of 

 of the dog, while in the rabbit either the mechanism of ingestion is in- 

 complete or the cells fail to. cope successfully with their contents after 

 ingestion. This example has been selected because anthrax bacilli have 

 been shown to contain less toxic intracellular bodies — endotoxins — 

 than many other infectious germs, and the likelihood of the rabbit being 

 poisoned by any primary plasma disintegration of the bacilli is not very 

 great, so that if the plasma mechanism had corresponded to that of the 

 serum the animal should have been saved. The validity of such an 

 argument would not have been so apparent if we had substituted 

 cholera vibrios for anthrax bacilli in rabbits, for the bodies liberated 

 from cholera bacilli at their disintegration are veiy toxic. 



