ESSAYS ON BACTERIOLOGY. 143 



or may have been overlooked. Sections of tissue to 

 be examined for tubercle bacilli are treated as follows : 

 Tbey are stained for about thirty minutes in carbol- 

 fuchsin solution, washed for about one minute in 5 

 per cent, sulphuric acid, then for the same time in al- 

 cohol; they are then counter-stained for three or four 

 minutes in weak methyl blue, again quickly washed 

 in alcohol, immersed till clear in oil of cloves, and are 

 ready to examine in Canada balsam. 



It is to be remembered that the successful identifi- 

 cation of the tubercle bacilli requires that the material 

 shall come from the seat of disease, shall contain the 

 bacteria, and that the examination shall have been 

 properly made. Failure of any of these requirements 

 will necessarily yield but negative results. Fortu- 

 nately, there' remain two other available tests, the one 

 the inoculation of susceptible animals, already re- 

 ferred to, the other the diagnostic injection of tuber- 

 cuhn. 



The tuberculin injection is, like the Widal test 

 for typhoid fever and the Mallein injection for the 

 detection of glanders, an illustration of indirect bac- 

 -teriological diagnosis. In 1890 Koch announced the 

 discovery, in the culture-products of tubercle bacilli, 

 of a substance which, being injected into animals or 

 men suffering from tuberculosis, would produce cer- 

 tain local and constitutional reactions and, as he then 

 thought, curative effects. Hopes thus raised in the 

 latter direction have not been fulfilled, for reasons 



