122 CLINICAL BACTERIOLOGY AND EftEMATOLOGY 



have been saturated with amboceptor, or sensitized, provide 

 a test for the presence of complement. Add corpuscles so 

 prepared to any specimen of serum, normal saline, etc., and 

 if they are dissolved it shows that the fluid contained com- 

 plement, or vice versa. This is of fundamental importance 

 in the Wassermann reaction. 



For the sake of brevity we will omit any reference to the 

 theoretical interpretation of the Wassermann reaction which 

 is, indeed, at present not properly understood and will 

 describe as simply as possible the essentials of the test. They 

 may be stated in a few words : If a specimen of serum from 

 a syphilitic person, containing complement, be incubated 

 after admixture with certain emulsions of lipoid substances, 

 the complement will disappear, whereas if serum from a non- 

 syphilitic person is treated in the same way it will not. 



The emulsion of fatty substances is usually but incorrectly 

 termed the "antigen." It is prepared in several ways, but the 

 following will be found satisfactory : Take human heart or 

 liver from the post-mortem room, or sheep's heart or liver; 

 mince them finely (freeing it from fat or fibrous tissue in the 

 case of the heart), and to each gramme of material add 3 c.c. 

 of absolute alcohol. Shake thoroughly, and let the mixture 

 stand twenty-four hours, shaking occasionally. Then heat in 

 a water-bath at 60 C. for one hour (this is not really essen- 

 tial), and again shake thoroughly. Allow to settle, and dilute 

 the clear supernatant fluid when required for use in the fol- 

 lowing way : Take 9 volumes of normal saline solution, 

 measuring them out by means of a pipette, and place them 

 in a small test-tube. Now take i volume of the alcoholic 

 antigen, and float it on the surface of the saline solution. You 

 will see a turbid zone (due to the precipitation of liquid sub- 

 stances dissolved in the alcohol) at the zone of contact. After 

 five minutes or so give it a very slight shake, so as partially 

 to mix the fluids together. After another interval repeat the 

 process, and ultimately stir them together. It is advisable 

 that the fluids should always be mixed for use in this way. 



In actual practice the alcoholic solution of antigen requires 

 careful testing and standardizing, but this is hardly necessary 

 for the present, and will be described later. 



EXPERIMENT i. Mix together i volume of fresh serum 

 from a normal (or non-syphilitic) person, and 4 volumes of 



