SEROLOGICAL METHODS 205 



similarity in structure of the two molecules. In either case, it is usually- 

 possible to remove cross-reacting antibodies by absorption. The 

 heterologous antigen is mixed with the serum, allowed to react, and 

 removed by centrifugation. Usually, considerable antibody for the 

 homologous strain remains, whereas antibody for the absorbing strain 

 can no longer be demonstrated. 



The technic has been widely employed for establishing the antigenic 

 structure of bacteria, and a classic example is found in the Kauffmann- 

 White antigenic classification of the genus Salmonella (Edwards and 

 Ewing, 1955). 



To secure bacteria for agglutinin absorption, cultures are grown on a 

 suitable agar medium. Since large quantities of cells are needed, the 

 agar is dispensed in Roux bottles or petri dishes. The cells are washed 

 from the agar with saline and killed by adding phenol to a concentration 

 of 1 per cent. Heat killing may be used instead if the antigen under 

 study is not heat labile or if it is necessary to destroy a heat-labile antigen. 



The cells are centrifuged from the saline and washed twice with addi- 

 tional saline in graduated centrifuge tubes. The time and speed of 

 centrifugation should be standardized to permit uniform packing. To 

 1 ml of packed cells is added 9 ml of antiserum. High-titered sera are 

 diluted 1:5 or 1:10. The cells and serum are mixed and incubated at 

 37°C for 2 hr mth intermittent shaking. Following this, the cells are 

 removed by centrifugation and the serum tested for agglutinins against 

 the absorbing strain. If agglutinins remain, the serum is absorbed again, 

 using a second 1-ml portion of packed cells and the same incubation. 

 When absorption is complete, the serum should fail to react with the 

 absorbing strain but still react with the homologous strain unless the two 

 cultures were antigenically identical. In setting up an absorption, one 

 should include a control serum without cells to be incubated under the 

 same conditions as the test. An absorption of the serum with the 

 homologous strain may be included also. 



Absorptions are not always reciprocal, and if two cultures are to be 

 compared, the specific antiserum for each should be absorbed with the 

 heterologous strain (Krumwiede eial., 1925). 



Absorption can also be conducted with soluble antigens. One method 

 is to add small amounts of antigen to an antiserum on successive days 

 until a precipitate no longer occurs with that antigen. Another method 

 is to locate the region of slight antigen excess, using the supernate test 

 described under ^'precipitation.'^ A volume of antiserum and antigen 

 calculated to give slight antigen excess is then mixed and incubated at 

 0-5°C until precipitation stops. The precipitate is removed by centrifu- 

 gation, and the supernate is tested with the absorbing antigen and the 

 homologous antigen. 



