Report on the Bacteriology of Water. 401 



in the blood) for a visible fermentation by the B. coli communis to 

 take place. The bubbles of gas are certainly independent of any 

 ingredients present in either the gelatine or in the peptone, for I 

 have found them to form also in agar-agar-peptone, and also in meat- 

 extract gelatine to which no peptone had been added.* The great 

 convenience of the test depends upon its involving only the use of a 

 medium which must invariably be at hand at all times in every 

 bacteriological laboratory, and also on its dispensing with the use of 

 an incubating temperature, whilst it has the farther advantage over 

 Dunbar's original broth-bubble test that the bubbles of gas being 

 fixed in the solid gelatine, the tubes can be examined at leisure even 

 days or weeks after inoculation, whilst with the broth-bubble test, if 

 the tubes are not examined at the right time, the fermentation may 

 have ceased ; besides, in the broth, of course, the bubbles are not 

 nearly so conspicuous. Extensive use has been made of this method 

 during the present investigation, and for rapidly and certainly dis- 

 tinguishing between the typhoid bacillus and the B. coli communis I 

 have found it unequalled ; on the other hand, it must be borne in 

 mind that it does not serve to distinguish between the B. coli communis 

 and many other fermenting organisms. 



A farther but less certain distinction which should also be em- 

 ployed for differentiating between the typhoid bacillus and the 

 B. coli communis is the so-called indol-reaction. This test is best ap- 

 plied in the following manner, as recommended by Kitasato : 



To 10 c.c. of the culture in ordinary alkaline peptone-broth of the 

 organism under examination, and which has been growing for 

 twenty-four hours in the incubator, add 1 c.c. of a solution of potas- 

 sium or sodium nitrite (containing 0'02 gram in 100 c.c.) and then a 

 few drops of concentrated sulphuric acid. If indol is present, a rose to 

 deep-red coloration is produced, depending on the interaction of nitrous 

 acid with indol to form nitrosoindol nitrate which is of red colour. 

 On applying this test to the B. coli communis an indol-reaction should 

 be obtained, whilst the typhoid bacillus gives invariably a negative 

 result. In practice I have found it advisable not to apply the indol- 

 test until the broth- culture has been forty-eight hours in the in- 

 cubator. 



Although the B. coli communis is generally supposed to give the 

 indol-reaction, this character would appear not to be so constant as 

 is commonly imagined. In my own experiments I have known one and 

 the same culture- series of the B. coli communis not to give the indol- 

 reaction at one time, and yet subsequently to become possessed of 

 this power, although I have not been able to determine the cause 

 which leads to the loss of the indol-producing capacity. Thinking 



* Dunbar found also that no bubbles were formed in a solution of peptone -with- 

 out meat extract. 



