296 INFECTION AND IMMUNITY 



resuming its activity when diluted to isotonicity with distilled water. 

 Removal of salt by dialysis or other means of globulin precipitation 

 divides the complement into two fractions, the globulin fraction and 

 the albumin fraction, neither of which will act alone, but which 

 together possess the properties of undivided complement. The 

 globulin fraction attaches directly to the sensitized cells and is there- 

 fore spoken of by German investigators as "mid-piece." The al- 

 bumin fraction acts upon the sensitized cells only after attachment 

 of the globulin fraction and is therefore spoken of as "end-piece." 

 It is seen, therefore, that a great many of the properties of alexin 

 make it seem rather likely that this substance is quite similar to 

 ferments in its action. 



The Fixation of Complement by Precipitates. It has been found 

 by Gengou 49 and confirmed by Moreschi, Gay, 50 and others, that 

 when the serum of an animal immunized with the serum of another 

 species or with a foreign albumin is mixed with a solution of the 

 substance used in the immunization, the precipitate formed will 

 remove complement from the mixture. In other words, precipitates 

 formed by the reaction of precipitin with its antigen will fix com- 

 plement. This is of great importance in complement-fixation tests ; 

 for because of insufficient washing, the blood cells used in producing 

 the hemolytic amboceptor, may, from the presence of serum, give 

 rise to a precipitin as well as a hemolysin. In the test done subse- 

 quently, a precipitin reaction may take place and by thus removing 

 complement may give a false result. The absorption of complement 

 by such precipitates takes place when the two reacting factors, the 

 precipitin and its antigen, are in dilution so high a visible precipi- 

 tate can not be observed. This fact, together with others too com- 

 plicated to be discussed in this place, have led us to the belief 

 that the so-called precipitins are true sensitizers, exerting toward 

 unformed proteins the same function that the so-called sensitizer or 

 amboceptor exerts toward cellular formed antigens. (See p. 293.) 



Quantitative Relationship Between Amboceptor and Complement. 

 Morgenroth and Sachs 51 have succeeded in showing that within 

 certain limits an inverse relationship exists between these two bodies. 

 If for a given quantity of red blood cells a certain quantity of 



49 Gengou, Ann. Past., 1902. 

 5(1 Gay, Cent. f. Bakt., I, xxix, 1905. 



51 Morgenroth nnd Sachs, ' ( Gesammel. Arb. f iir Immimitatsf orschung. ' ' Berlin, 

 Hirschwald, 1904. 



