DIFFERENTIATING COMPLEMENTS. 223 



At least this had been the case thus far, for a partial anticomplement, 

 one acting only against a single complement, had not been observed. 



Through the courtesy of Dr. Cnyrim we detained a normal anti- 

 complement which possessed the desired properties, and we there- 

 fore gladly availed ourselves of this favorable opportunity to demon- 

 strate, by means of the elective binding of anticomplement, the 

 difference between two complements in one and the same serum, a 

 difference that had not heretofore been demonstrated. 1 



This normal anticomplement was an ascitic fluid derived from a 

 case of cirrhosis of the liver; it exerted a marked antihsemolytic 

 action in one particular case. By means of an experiment we first 

 determined that this action was due to the presence of an anticomple- 

 ment and not of an anti-immune body. This showed us that the 

 ascitic fluid exerted practically no influence on the anchoring of the 

 immune body in question to the red blood-cells. 



The serum whose complements we examined was guinea-pig 

 serum, which activated two amboceptors obtained by immunization. 

 These amboceptors were contained in the inactive serum of a rabbit, 

 A, which had been immunized with ox blood; and in the inactive 

 serum of a goat, B, which had been immunized with sheep blood. 

 Corresponding to this, ox blood w r as used for case A, and sheep blood 

 for case B. The inactive ascitic fluid does not dissolve these species 

 of blood even after the addition of guinea-pig serum 



To begin, we saturated ox blood-cells with the specific amboceptor 

 by adding 0.01 cc. immune body A to each 1 cc. of a 5% suspension 

 of the cells. This is about ten times the amount which on the addi- 

 tion of sufficient complement (0.1 cc. guinea-pig serum) effected 

 complete solution. The mixture was placed in the incubator and 

 frequently shaken. At the end of one hour it was centrifuged, the 

 fluid poured off, and the blood-cells, loaded with amboceptor, sus- 

 pended in salt solution. In exactly the same manner sheep blood- 

 cells were treated with the inactive serum B, 0.2 cc. for each 1 cc. 

 of the 5% suspension. On the addition of guinea-pig serum to these 

 blood-cells, haemolysis ensued very quickly in the thermostat; in 

 both cases it required 0.008 cc. guinea-pig serum to fully dissolve 1 cc. of 

 the suspension, while 0.0065 cc. caused incomplete solution and 0.002 



1 In the following, for the sake of simplicity, we shall speak only of two 

 complements, whereas we wish here to remark that two groups of complements 

 are probably to be understood, each group made up of a host of single comple- 

 ments which it is impossible thus far to analyze. 



