EEACTIONS OF THE COLI-TYPHOID BACILLI 391 



manifest agglutinability at all. Thus the only criteria for preferring a 

 strain are that it is agglutinated by more sera than the other strains 

 available or that it is a sub-culture of what may be termed a historic 

 strain. Even with the most carefully selected strain, however, a serum 

 which agglutinates it may have no effect on an organism which, accord- 

 ing to the other available evidence, ought to be identified as belonging to 

 the species to which the selected strain belongs. These remarks apply 

 with special cogency to the paratyphoid bacilli and to Flexner's dysentery 

 bacillus— strains of both of which types appear to possess much greater 

 individuality than is the case with the typhoid bacillus on the one hand 

 or with Shiga's dysentery bacillus on the other. Given a typical strain, 

 it is to be noted that successive cultures may show differences in agglutin- 

 ability. Thus the culture media used should be as uniform as possible 

 in composition. The best medium for growing cultures for agglutination 

 tests is probably a pale veal bouillon which has not been too long auto- 

 claved, and it is well to make daily cultures for from fpur to six weeks 

 on this medium. After such preparation a 24 to 48 hours' culture 

 may be used as the emulsion for agglutination observations, though 

 many observers prefer to use agar slope cultures washed off with saline or 

 bouillon. Though living bacilli may be more readily agglutinated than 

 dead organisms there is great convenience in using killed cultures as 

 recommended by Dreyer (p. 118), greater uniformity being thereby 

 attained. It is of great importance for routine work and especially when 

 successive observations on the same case are to be made that the 

 emulsions used should contain approximately the same number of bacilli 

 per unit of volume. Ledingham recommends an emulsion of about 2000 

 million bacilli per c.c. ; if the degree of opacity of such a concentration 

 be observed, successive emulsions can be readily standardised. It is here 

 that killed cultures are specially convenient, as large batches of these can 

 be easily prepared. With regard to the time and temperature factors in 

 an agglutination reaction, when relatively high concentrations of serum 

 are being used, a full effect may be obtained in from £-1 hour at room 

 temperature especially with the microscopic method ; with low con- 

 centrations (here it is advisable to practise the sedimentation technique) 

 the mixtures should be kept in a water bath for two hours at 55° C. 



The diagnosis of the nature of an intestinal infection by the 

 agglutination method is relatively simple if the serum agglutin- 

 ates only one species of bacterium. The chief point to be borne 

 in mind here is that sometimes the serum of a normal individual 

 may agglutinate an organism. Such" a phenomenon, however, 

 is usually only met with when the serum is in strong con- 

 centration, and the difficulty can be overcome by employing 

 dilute solutions. Thus in suspected typhoid fever if the patient's 

 serum in a dilution of 1-30 of bouillon or normal saline 

 agglutinates the typhoid bacillus a positive diagnosis may be 

 given. In paratyphoid A infections a diagnosis may be founded 

 on agglutination with a dilution of 1-10, in paratyphoid B 

 infections with a dilution of 1-25, and in Shiga dysentery 

 infections with a dilution of 1-50. In the case of Flexner's 

 dysentery bacillus agglutination may occur with normal serum 



