NON-SPECIFIC COMPLEMENT FIXATION 397 



of hemolysis, i. e., if a small amount of complement is absorbed, hemol- 

 ysis will be correspondingly incomplete, if all complement is absorbed, 

 hemolysis does not occur at all. If, therefore, in any complement-fixa- 

 tion test the antigen is used in an amount that will give this non-specific 

 absorption, even to a slight degree, a grave source of error is introduced. 



As a general rule for all complement-fixation tests, the dose of antigen 

 employed should never be more than one-fourth or one-half of its 

 anticomplementary dose (that amount which of itself is capable of 

 non-specific complement-fixation). 



2. A serum may of itself absorb a small amount of complement, espe- 

 cially if it is old or infected with bacteria. This is known popularly as 

 the anticomplementary action of a serum, and in every complement- 

 fixation test in order to detect this condition a proper control, consisting 

 of the dose of serum used plus complement, hemolysin, and corpuscles 

 is required, the non-specific absorption of complement being determined 

 by the results of hemolysis. 



Moreover, perfectly fresh serums may show this non-specific absorp- 

 tion of complement to a slight degree, especially in the presence of the 

 lipoidal extracts used as "antigens" in the Wassermann reaction. 



Heating a serum to 56 C. for half an hour largely removes this 

 anticomplementary effect of serums, unless they are quite old and in- 

 fected; accordingly, heated serums are used almost exclusively in com- 

 plement-fixation tests. This is usually called inactivation, or the removal 

 of native complement from a serum, but in the majority of instances 

 the complement of a serum generally deteriorates rapidly, and the serum 

 is heated mainly for the purpose of removing its anticomplementary 

 action, i. e., its ability to effect non-specific absorption of complement. 



By referring to the original Bordet experiment, it will be observed 

 that this investigator controlled any non-specific absorption of comple- 

 ment by both the immune and the normal serum in tubes C and D of 

 the series by using the full dose of these serums, with a similar amount of 

 complement, and noting that hemolysis was complete. His controls, 

 E and F, were to determine if the process of inactivation or removal of 

 native complement from the two serums was complete, and the total ab- 

 sence of hemolysis showed that it was. His control on the anticomple- 

 mentary action of the antigen was also included in tube D, for if the 

 emulsion alone had absorbed complement to any degree, hemolysis 

 would have been incomplete. 



Quantitative Factors in Complement-fixation Tests. From what 

 has been said it will, therefore, readily be appreciated that complement- 



