TNDOL-FORMATION BY BACTERIA 85 



nitrites as well as indol, but usually in making the test (e.g., in 

 the case of b. coli) the nitrites must be added. This is effected 

 by adding to an ordinary tube of medium 1 c.c. of a '02 per 

 cent, solution of potassium nitrite, and testing with pure nitric 

 or sulphuric acid. In any case only a drop of the acid need be 

 added to, say, 10 c.c. of medium. If no result be obtained at 

 once it is well to allow the tube to stand for an hour, as some- 

 times the reaction is very slowly produced. In many instances 

 incubation at 37 C. for several days may be necessary before 

 the presence of indol is demonstrable. The amount of indol 

 produced by a bacterium seems to vary very much with certain 

 unknown qualities of the peptone. It is well, therefore, to test 

 a series of peptones with an organism (such as the b. coli) 

 known to produce indol, and, noting the sample with which the 

 best reaction is obtained, to reserve it for making media to be 

 used for the detection of this product. This method has for 

 long been felt not to be satisfactory, and the following at present 

 bids fair to replace it : 



(2) Ehrlictis Rosindol Reaction. The adaptation of this to 

 bacteriological purposes was brought forward by Bohme in 1906. 

 For ease of application and delicacy of effect the reaction 

 possesses great advantages. It depends on the fact that 

 paradimethylamidobenzaldehyde unites with indol to form a 

 rosindol body whose colour is readily developed, especially in 

 presence of an oxidising substance such as potassium per- 

 sulphate (K 2 S 2 O S ). Two solutions are required : 



(1) Paradimethylamidobenzaldehyde (Griibler) 4 grms. 

 Absolute alcohol (96 per cent.) . . . 380 c.c. 

 Concentrated hydrochloric acid . . 80 c.c. 



(2) Potassium persulphate . Saturated watery solution. 



To a 10 c.c. bouillon culture of the organism add 5 c.c. of (1) 

 and then 5 c.c. of (2), and shake well (according to MacConkey 

 1 c.c. of each solution is sufficient) ; if indol be present a rose- 

 red colour will appear in a few minutes. Sometimes the rose 

 colour appears on the addition of solution (1), and the addition 

 of a special oxidising agent is unnecessary. The rosindol com- 

 pound can be separated from the culture by shaking the latter 

 up with ainyl alcohol, and MacConkey recommends that this 

 should be done in cases of a doubtful reaction, as sometimes 

 when a faint pink colour appears in the culture tube the 

 extracting alcohol remains colourless, showing th&t^no real 

 reaction has occurred. Marshall has pointed out that by x means 



