CONCERNING ALEXIN ACTION. 191 



method can lead to the greatest errors. In this respect a comparison 

 of the results obtained by Buchner and Gruber, is very instructive, 

 for among their cases are two combinations which are designated 

 by the one as positive, and by the other negative. 



The amboceptor of rabbit serum for sheep blood, which Buchner, 

 because of the failure to reactivate this with sheep serum, regarded 

 as absent, Gruber, by means of the cold separation method, could 

 demonstrate as present; and for ox serum, whose amboceptor Buchner 

 had already demonstrated by the activation with guinea-pig serum, 

 Gruber, through the failure of his cold absorption method, arrived 

 at the view of a pure alexin action. 



In Gruber's negative cases, which embrace the following com- 

 binations: I, rabbit blood dog serum; II, rabbit blood ox serum; 

 III, gui::ea-pig blood ox serum, IV, rabbit blood guinea-pig serum; 

 I have systematically sought for sources of fitting complements 

 and have found these in abundance. Naturally in view of the experi- 

 ences above mentioned the inactivation of the sera was effected 

 at the lowest temperatures possible; thus dog serum and guinea-pig 

 serum at 50, ox serum at 52 C. In the following sera (in part in 

 agreement with other previous experiences) I have found comple- 

 ments suitable for activation: 



I. For the amboceptor of dog serum, acting on rabbit blood; in guinea- 

 pig serum, ox serum, goat serum, and sheep serum. 



II. For the amboceptor of ox serum, acting on rabbit blood; in 

 guinea-pig serum, rabbit serum, and rat serum. 



III. For the amboceptor of ox serum, acting on guinea-pig blood; 

 in guinea-pig serum, human serum, rat serum, horse serum, and to 

 a slight extent also in sheep serum 



Naturally in all the experiments, control tests were made with 

 the active serum, which served as complement. In the cases desig- 

 nated as positive completion, this serum by itself had to exert no 

 haemolytic action or at least to act in a very much smaller degree. 



Gruber's fourth negative case, rabbit blood and guinea-pig serum, 

 offered considerable difficulties because the combination is very 

 little or not at all effective, and it is probably because of this that 

 Gruber speaks of " concentrated guinea-pig serum." Among a large 

 number of guinea-pig sera examined for this purpose, we found 

 only two sufficiently hsemolytically active. But here also, through 

 the successful activation by means of human and ox sera (sera, to 

 be sure, which by themselves dissolve rabbit blood, but which still 



