152 INFECTION AND IMMUNITY. 



conclusions regarding the importance of phagocy- 

 tosis purely from a study of the process in vivo. In 

 spite of this we have come to a better understand- 

 ing of the mechanism of phagocytosis by studies 

 of the phenomenon under artificial conditions, as 

 will appear later. 



In determining the germicidal action of serum. 



Action of , . , , T -, , , . T 



serum, which should be freshly obtained, it may be 

 mixed with a suspension of the microbes in a 

 number of test-tubes, varying amounts of serum 

 being used with constant amounts of the bacteria 

 in the different tubes. At a subsequent period, 

 from three to twenty-four hours later, cultures on 

 Petri plates are made from these mixtures. The 

 numbers of colonies which appear in these cultures, 

 compared with the number which appear when 

 serum is not. added, is an index of the bactericidal 

 power of the serum. If this power is found to be 

 high, it is, in the present state of our knowledge, 

 considered as presumptive evidence that the nat- 

 ural immunity of the animal depends on it, at 

 least in part. It is, nevertheless, a fact that the 

 antibacterial immunity of an animal does not 

 always go hand in hand with the bactericidal power 

 of its serum. A well-known illustration of this is 

 the following: Both the dog and the rat have a 

 rather high degree of immunity against infections 

 with the anthrax bacillus; yet it has been found 

 that the serum of the dog has almost no bacteri- 

 cidal effect on this microbe, while that of the rat 

 has a very strong effect. At the same time we 

 should remember that the bactericidal power of the 

 serum does not necessarily represent the entire 

 antibacterial function of the body. In the serum 

 we have none of the body cells, and especially none 



