THE AGGLUTINATION REACTION 309 



37 C. for two hours, when the results are read. A control of bacterial 

 emulsion and salt solution should be included, and it is advisable to 

 test out the immune sera from time to time with pure cultures of the 

 homologous strains. 



5. Microscopic tests are conducted by mixing on cover-slides plati- 

 num loopsful of bacterial emulsion and undiluted and diluted immune 

 sera. These slides are then suspended as hanging-drop preparations 

 and the results read after one-half to an hour. A control should always 

 be included. 



TECHNIC OF THE ABSORPTION AGGLUTINATION TEST IN MIXED INFECTION 

 (THE SATURATION TEST OF CASTELLANI) 



The practical importance of partial agglutinins is recognized in the 

 diagnosis of mixed infections. Thus the serum of a patient may agglu- 

 tinate typhoid as well as paratyphoid bacilli in dilutions up to 1 : 100. 

 This may indicate one of three possibilities: 



1. The patient may be infected with typhoid, but has formed an 

 exceptionally large quantity of group agglutinins for paratyphoid bacilli. 

 Saturation of this serum with typhoid bacilli will remove all the typhoid 

 and a portion, if not all, of the group agglutinins. Saturation with para- 

 typhoid bacilli will remove the group agglutinins, but not the main or 

 typhoid agglutinin. 



2. The patient may be infected with paratyphoid bacilli, but has 

 formed, at the same time, many partial agglutinins for typhoid bacilli. 

 Saturation of the serum with paratyphoid bacilli will remove all the 

 paratyphoid and a large portion of the typhoid agglutinin. 



3. The patient may have a mixed infection of typhoid and para- 

 typhoid, and therefore agglutinin for both may be present. Saturation 

 of the serum with typhoid bacilli will remove the typhoid and probably 

 a small portion of the paratyphoid agglutinin. After this reaction the 

 serum will still show the presence of a decided quantity of paratyphoid 

 agglutinin. 



In selecting the most likely one of these hypotheses a decision may be 

 reached by adopting the method of Castellani (Citron), which is as fol- 

 lows: 



1. Four rows of test-tubes are arranged, each row being made up of 

 four small tubes each containing 1 c.c. of serum dilutions 1 : 20, 1 : 40, 

 1 : 80, and 1 : 160 respectively. 



2. In each of the tubes of the first and second rows five loopfuls of 

 typhoid bacilli are emulsified. An extra tube containing 1 c.c. of nor- 



