118 PROPERTIES OF ILEMOLYTIC SERA 



employed the method of injecting the fluid and stromata 

 respectively into animals, and observing whether or not 

 immune-body was developed. Bordet alone obtained con- 

 firmatory results by test-tube experiments, though he does 

 not give details as to the exact amounts. It has seemed 

 to us desirable to investigate the question in this latter 

 way, as the experiments can be much more rapidly and 

 exactly carried out. And, further, if the method of animal 

 injection is to be employed, the blood of the species of animal 

 injected ought normally to have no hsemolytic action on the 

 corpuscles used. The serum of the guinea-pig produces 

 a certain amount of lysis of rabbit's corpuscles, and this 

 varies much in the case of different animals : there would 

 accordingly be great liability to error unless the haemolytic 

 action present after injection of stromata or fluid were of 

 pronounced character. It has now been demonstrated so 

 frequently that when a substance injected into the body 

 of an animal gives rise to the production of an anti-substance 

 there is a specific combining affinity between the substance 

 and the anti-substance, that this maybe accepted as a general 

 law. In the case before us the question thus comes to 

 be, does the immune-body combine with molecules in the 

 ' stromata ' or in the ' contained fluid ' of the red corpuscles ; 

 or, in other words, when the corpuscles undergo lysis, do the 

 receptors remain in the stromata or do they pass out into 

 the surrounding fluid ? It might appear that an answer 

 to this question could be obtained by adding immune-body 

 to the stromata and separated fluid respectively, and 

 observing whether or not it was taken up. As, however, 

 it has been shown that the combination of immune-body 

 with receptors is a very loose one (vide p. 12) in fact, 

 belongs to the class of ' reversible actions ' it follows 

 that even if immune-body did enter into combination, a 

 certain amount would be obtainable by dissociation, and 

 thus accurate knowledge as to the combination or non- 



