OUTLOOK. 689 



salt solution, we shall find that within a short time a solution of the red 

 blood-corpuscles takes place. The solution becomes transparent and a 

 clear red. If exactly the same experiment is carried out with inactive 

 serum, the red corpuscles will remain unaffected. Now if in a third test- 

 tube the serum of a normal guinea pig is added to the suspension of 

 blood-corpuscles, there is again no hemolysis. On the other hand, the 

 solvent effect is obtained if the normal serum from normal guinea pigs is 

 added to the inactivated serum. This proves that at least two substances 

 are necessary for bringing about hemolysis. One of these is found 

 already formed in the immune serum and also in normal serum, and the 

 other is only yielded by immunized serum, i.e., in this case in the blood 

 of guinea pigs which have been previously treated with the blood 

 of rabbits. The two substances are different as regards their behavior 

 toward heat. That which is present in normal serum is stable towards 

 heat, while that present in the other is unstable. Quite a number of different 

 names have been given to these substances. We shall choose for the 

 thermo-stable substance the designation amboceptor, and for the thermo- 

 unstable one the name complement. 



Ehrlich's theory fits in at this point. He was especially interested in 

 explaining why the different sera should act so specifically, and what the 

 relations of amboceptor and complement are to the red blood-corpuscles 

 and to themselves. For simplicity's sake we will designate the two com- 

 ponents, amboceptor and complement, which produce hemolysis, by the 

 name hemolysin. This must have, judging from its analogy to the toxines, 

 a very distinct affinity to some constituent of the red corpuscles. Ehrlich 

 and Morgenroth's experiments have proved this. They injected sheep's 

 blood into a goat. The serum from this animal completely dissolved the 

 sheep-blood corpuscles, although it lost this property on heating to 55 

 degrees. The complements were destroyed. The amboceptors must 

 have remained unchanged, as they are thermo-stable at this temperature. 

 The above investigators then added sheep-blood corpuscles, and centri- 

 fugalized the mixture after standing half an hour. To the resulting serum 

 they added more sheep-blood corpuscles, and also some fresh normal 

 serum. No hemolysis resulted. When the previous addition of sheep- 

 blood corpuscles was omitted, and normal serum added to the inactivated 

 serum, the solution of the red corpuscles immediately set in. It follows 

 from these experiments that the sheep-blood corpuscles had removed one 

 of the components of the hemolysin which was, in fact, the amboceptor. 

 That this assumption is correct, is evident from the fact that the blood- 

 corpuscles, centrifugalized as above, immediately went into solution on 

 the addition of normal serum in an 0.85 per cent salt solution. We must 

 also mention that normal goafs-blood serum does not attack the normal 

 sheep-blood corpuscles. 



