ALEXIN ABSORPTION. 409 



denced by this experiment, the antagonistic property inhibits the 

 hemolysis produced by a normal sensitizer.* 



Although the antagonistic power prevents hemolysis in a mixture 

 of bovine corpuscles, normal rabbit serum, 56 degrees, and guinea- 

 pig alexin, because the serum is only faintly sensitizing for the 

 corpuscles, there is no inhibition when goat corpuscles are used, be- 

 cause they are powerfully sensitized by the rabbit serum. In the 

 latter case, as we have just seen, the antagonistic property simply 

 retards hemolysis. This effect may be shown in another way, as 

 is evidenced by the experiments of Pfeiffer and Friedberger and of 

 Sachs. In their experiments the property that opposes the antag- 

 onistic property, namely, the sensitizer, is removed. After treating 

 normal rabbit serum (56 degrees) with a sufficient amount of goat 

 corpuscles we have a fluid that is no longer sensitizing, but purely 

 antagonistic; it will therefore protect moderately sensitized goat 

 corpuscles from hemolysis. 



If the sensitizer (whether in an immune serum or normal serum) 

 is too strong, the antagonistic property will be overcome and hemoly- 

 sis take place. The origin of the sensitizer is immaterial, but its 

 strength is all-important. 



This fact is quite conceivable in view of the fact that the sensitizer 

 does not unite directly with the alexin. It is the sensitized cor- 

 puscle that unites with the alexin, and the greater the sensitization 

 the better this union. The union with the sensitizer changes the 

 properties of molecular adhesion in the corpuscle so that its avidity 

 for the alexin is increased ; in a similar manner the action of agglu- 

 tinins on bacteria is to increase their reaction to the clumping 

 effect of electrolytes. Any result produced by the sensitizer depends, 

 not on its proper nature, but on its effect. 



* This antagonistic effect is so distinct that goat corpuscles that have been 

 sensitized by a small dose of rabbit serum and then suspended in salt solution are 

 hemolyzed by alexin more rapidly than the corpuscles sensitized by twice as much 

 serum when kept in this serum. For example: in each of four tubes is placed 

 0.05 c.c. of goat blood; to A and B is added 0.4 c.c. of normal rabbit serum, 

 56 degrees and to C and D 0.2 c.c. of the same serum. After 2 hours' contact 

 A and C are centrifugalized, the supernatant fluids removed and 0.4 and 0.2 c.c. 

 of salt solution added to the sediments in A and C respectively. To each of the 

 four tubes is then added 0.1 c.c. of guinea-pig alexin. Hemolysis is complete in 

 A in 1 5 minutes, in C in a half hour, in B in 50 minutes, and in D in a little over 

 an hour, 



