218 MANUAL OF MICROBIOLOGICAL METHODS 



antigen to exert any possible anticomplementary effect. Following the 

 initial incubation, place 1.0 ml of sensitized erythrocytes in each tube, 

 mix, and incubate in a 37°C water bath for 30 min. 



Since the optimal antigen dilution will not be known at the outset, the 

 titration may be performed as above, substituting an additional 0.5 ml of 

 saline for the antigen. In either procedure, a control consisting of 1.0 ml 

 of sensitized erythrocytes and 2.0 ml of saline should be included. The 

 tubes are read by comparison with this control, which, of course, should 

 display no lysis. Complete lysis is reported only in those tubes with no 

 trace of turbidity. The smallest amount of complement yielding complete 

 hemolysis is called the exact unit; however, the unit actually employed is 

 called the full unit which is contained in the next higher tube, or the tube 

 with 0.05 ml more complement. In the actual test, two full units are 

 added to each tube. Thus, if the exact unit were contained in 0.35 ml, the 

 full unit would be 0.40 ml and 0.8 ml would be the volume employed in 

 the complement-fixation test. It is usually more convenient to calculate 

 a complement dilution such that two full units are contained in 1 ml. 

 This is done by dividing the reciprocal of the dilution by the volume con- 

 taining two full units. In the example given, 



1^ = 37.5 or 1 ml of 1:37 

 U.o 



The complement units as determined by the foregoing (Kolmer) pro- 

 cedure are based on lysis of 100 per cent of the cells and are sufficiently 

 precise for most practical applications. Some workers prefer to use a 

 50 per cent end point in calculating the unit, since the 50 per cent unit 

 may be determined more accurately. A description of this procedure 

 may be found in Kabat and Mayer (1948). 



Antigens. Microbial antigens used for complement-fixation tests are 

 often suspensions of washed, whole cells prepared like antigens for the 

 agglutination test. It is not uncommon to find that such antigens possess 

 some anticomplementary activity ; however, this can usually be removed 

 by dilution. Special procedures have been devised for preparing certain 

 microbial antigens used in serodiagnosis (Kolmer et al., 1951). In addi- 

 tion to whole cells, one may employ a cell lysate or mechanically disrupted 

 cells as antigen, and where possible, the use of a purified antigen should be 

 considered. Certain antigens for the complement-fixation test, mostly 

 viral and rickettsial, can be purchased from biological houses. 



Antigens are standardized and preserved by procedures described 

 under ''Agglutination" and ''Precipitation." Special procedures for 

 viral antigens will be found in the chapter on viruses. 



Antigens suitable for use in the complement-fixation test must not be 

 hemolytic or anticomplementary within the range where they have suffi- 



