ESSAYS ON BACTERIOLOGY. 139 



quired varies from a few minutes to half an hour or 

 more, the bacteria-containing fluid being meantime 

 protected against drying by vaselin or other adhesive 

 appKed to the margin of the cover glass. 



The specific character of this Widal reaction is still 

 under debate. Should it prove to be such, it will be 

 of great clinical value in this disease, and may open 

 the way to a similar diagnosis of other diseases whose 

 bacteria are known but are difficult of identification by 

 ordinary means. 



A method of diagnosis closely associated with, 

 though neither originating from nor entirely depend- 

 ent upon bacteriology, is that by inoculation. Inocu- 

 lation is frequently practiced as the final step in an 

 otherwise incomplete bacteriological demonstration, 

 but is sometimes sufficient of itself, and directly-, to 

 establish a diagnosis. If in a suspicious case the dis- 

 covery or identification of the bacteria be difficult or 

 impossible, inoculation of animals with material from 

 the suspected body may result in such an unmistak- 

 able attack of the disease as to remove all doubt. Or 

 such inoculation, with its resulting disease in the ani- 

 mal, may afford an easy opportunity for the discovery 

 and identification of the bacteria. Thus glanders, 

 rabies and tetanus in animals or human beings may 

 usually be certainly identified by inoculation of sus- 

 ceptible aj)imals. By this method obscure cases of 

 tuberculosis, in which the bacilli are not easily found, 

 , , such as those involving the skin, glands, bones, joints, 



