CONCERNING ALEXIN ACTION. 189 



bear slight warming, in order to remain capable of reacting, we had 

 to abandon our custom of inactivating sera by simply heating them 

 to 60 C. Thereafter we had always first to determine the minimal 

 inactivating temperature for each individual case. The limits of 

 temperature can usually be determined accurately; for dog serum 

 it is 49 C. We have also tried by means of other complements to 

 activate dog serum inactivated at 50, and have found a suitable 

 complement not only in guinea-pig serum but also in human serum. 

 In this case also, the thermolability of the amboceptor showed itself, 

 for heating to 60 C. destroyed the reactivatibility. In two cases, 

 however, the power to reactivate was preserved to a greater or less 

 extent even after heating to 60 C. In like manner dog serum 

 could be activated by the complements described when it had been 

 deprived of its solvent power by other means. Thus the comple- 

 ments of dog serum were absorbed by means of yeast, and by means 

 of an anticomplement serum (from a goat) whose normal amboceptor 

 for guinea-pig blood had been removed by washing with guinea-pig 

 blood. The dog sera so inactivated manifested their amboceptor 

 properties when they were appropriately activated. 



In the first two negative cases of Buchner, separation in the cold 

 had shown the presence of amboceptors in the sheep and rabbit serum. 

 I now sought by means of activating experiments to find fitting 

 complements for these amboceptors in other sera. Naturally, after 

 the above experiences, it was necessary here also to first determine 

 the minimal inactivating temperature. For sheep serum this is 

 50 C., for rabbit serum 51 C. If sheep serum is inactivated by 

 half an hour's heating to 50 C., it is easy to restore the hsemolytic 

 action on guinea-pig blood (Buchner's Case I) by the addition of 

 fresh human serum. In this way the complex nature of the normal 

 haemolysin of sheep serum can be demonstrated. One can also acti- 

 vate with guinea-pig serum, although then a feebler solvent action 

 is obtained. In both cases the thermolability of the amboceptor 

 is readily demonstrated; for by heating the sheep serum to 60 C. 

 this can no longer be activated, or only in very much less degree. 1 



1 in addition to this I have also demonstrated a thermolability of the am- 

 boceptors of goat serum which are activated by horse serum and act on 

 rabbit and guinea-pig blood. Repeated investigations by Dr. Morgenroth 

 have shown that a markedly thermolabile amboceptor is contained also in horse 

 serum. This amboceptor, which fits guinea-pig blood, and can no longer be 



