DIAGNOSIS OF TYPHOID FEVER AND FOOD-POISONING INFECTIONS 113 



The first three injections are given at intervals of 48 hours. 



Several days after the third injection o.i cc. of a i6-hour-old bouillon culture, 

 unheated, is given and several days later 0.5 cc. of a i6-hour-old bouillon culture, 

 unheated, is given. Ten days later the animals are bled and the blood serum 

 collected. 



If the procedure has been successful, it will be discovered by agglutination 

 tests that this specific immune serum in high dilutions (1:20,000 to 1:50,000) 

 will agglutinate typhoid bacilli. It will not agglutinate colon, paratyphiod A., 

 paratyphoid B, enteritidis of Gaertner nor other bacilli at all, or only when 

 much less diluted (1:100 to 1:5000). 



By immunizing animals in a like manner with paratyphoid A bacillus a 

 specific agglutinating serum for this organism can be obtained; by immunizing 

 with paratyphoid B bacillus, agglutinins specific for it may be obtained. Like- 

 wise specific agglutinins for most of the members of the typhoid-colon group of 

 bacteria may be produced. 



When an organism of the typhoid-colon group is isolated from a patient's 

 blood, urine, feces, or is found in water or food its nature may be determined as 

 follows: 



Make a 24-hour-old bouillon culture of the organism; make dilutions of a 

 specific typhoid serum from i : 100 to the titre limit of the serum. With the sus- 

 pected culture and known serum make quantitative agglutination tests. If the 

 organism is the typhoid bacillus it will be agglutinated in dilutions up to the 

 titre limit of the serum. If it is not the typhoid bacillus but some other member 

 of the typhoid-colon group it will not be agglutinated in any dilution or only in 

 dilutions much lower than the dilutions in which the same serum agglutinated 

 the typhoid bacillus. 



By repeating the experiment with the same culture and a specific paraty- 

 phoid A serum, and if necessary again with paratyphoid B serum, eventually we 

 will find a specific serum which agglutinates the organism in the same high dilu- 

 tion as it agglutinates the organism with which the specific serum was produced 

 and the classification is then established. 



VALUE OF AGGLUTINATION TESTS 



Agglutination tests made with undiluted serum or diluted less than i : 20 are 

 without significance for two reasons: 



1 . Under such conditions some sera exert an antiagglutinating effect. 



2. The sera of some apparently normal non-immune individuals under such 

 conditions will agglutinate one or more of the members of the typhoid-colon 

 group. 



The Widal test, made in dilutions of from i : 20 to i : 100, in suspected typhoid 

 fever cases gives accurate results in more than 75 per cent., ^ome claim in more 

 than 90 per cent, of cases; negative if the disease is not typhoid; positive if 

 the disease is typhoid. 



The sources of error, other than bad technique, in the Widal test, are as 

 follows: 



