560 THE RELATION OF COLLOIDS AND LIPOIDS TO IMMUNITY 



change is not yet understood. The reaction is based upon the following 

 generalizations : 



1. Solutions containing colloids i. e., antigen alone, antiserum alone, 

 or antigen plus non-specific antiserum in certain dilutions act in the 

 foregoing system by shifting the phenolphthalein end-point (the point of 

 neutralization when acid and alkali are brought together in the presence 

 of this indicator) in the sense of increased OH-ions (pink color). 



2. Specific antigens can inhibit the activity of their specific antise- 

 rums, the specific antigen-antibody combination then becoming evident 

 in vitro by a shift of the end-point in the sense of increased H-ion con- 

 centration (light color). 



Specificity. The specificity of the reaction has been confirmed by a 

 number of investigators who used the test for the identification of a 

 host of antigen-antibody combinations in vitro. The underlying prin- 

 ciples have been confirmed by Kraus and Amiradzibi, 1 Schroen, 2 

 Seifert, 3 Mosbacher, 4 and others. The reaction has been applied to a 

 study of various antigens and their antibodies, such as diphtheria toxin, 

 tetanus toxin, typhoid and tubercle bacilli, tumor extracts and placenta 

 extracts by Weichardt; extracts of syphilitic livers and serums of 

 syphilitic patients by Seifert, Keidel and Hurwitz. 5 



Technic. The technic of this reaction has been modified from time 

 to time. The method here given is essentially the latest given by 

 Weichardt, 6 slightly modified by Keidel and Hurwitz. 



Five constituents enter into the test: 



1. The Antigen. This is an alcoholic extract of syphilitic liver, 

 prepared in exactly the same manner as for performing the Wassermann 

 reaction. High dilutions of the antigen, ranging from 1 : 100 to 1 : 10,000, 

 are prepared with normal salt solution. As in the Wassermann re- 

 action, not every antigen is satisfactory, a point that can be determined 

 only by making preliminary tests. 



2. The patient's serum should be fresh, unheated, and highly diluted, 

 the dilutions ranging from 1 : 100 to 1 : 10,000,000. Usually it is better to 

 use higher than lower dilutions. When too concentrated solutions of 

 serums and of antigen are used, erroneous results are likely to be ob- 

 tained. 



3. A normal solution of sulphuric add. 



1 Zeitschr. f. Immunitatsforsch., 1910, vi, 16. 



2 Munch, med. Wochenschr., 1910, 38, 1981. 



3 Deutsch. med. Wochenschr., 1910, 50, 2333. 



4 Deutsch. med. Wochenschr., 1911, 22, 1021. 

 6 Jour. Amer. Med. Assoc., 1912, lix, 1257. 



8 Berl. klin. Wochenschr., 1911, 43, 1935. 



