METHODS OF EXAMINATION 345 



salt neutral-red agar, or in the medium of Drigalski and Conradi 

 (v. p. 42). After that period, though the continued infectiveness 

 of the disease indicates that they are still present, their isolation is 

 very difficult. We have seen that after ulceration is fairly estab- 

 lished by the sloughing of the necrosed tissue, the numbers present 

 in the patches are much diminished, and therefore there are fewer 

 cast off into the intestinal lumen, and that in addition there is a 

 correspondingly great increase of the b. coli, which thus causes 

 any typhoid bacilli in a plate to be quite outgrown. From the 

 fact that the ulcers in a case of typhoid may be very few in 

 number, it is evident that there may be at no time very many 

 typhoid bacilli in the intestine. The microscopic examination 

 of the stools is of course useless as a means of diagnosing the 

 presence of the typhoid bacillus. 



Isolation from Water Supplies. A great deal of work has been done 

 on this subject. It is evident that if it is difficult to isolate the bacilli 

 from the stools it must a fortiori be much more difficult to do so when 

 the latter are enormously diluted by water. The b. typhosus has, how- 

 ever, been isolated from water during epidemics. This was done by Klein 

 in the outbreaks in recent years at Worthing and Rotherham. The b. 

 coli is, as might be expected, the organism most commonly isolated in 

 such circumstances. In the case of both bacteria, the whole series of 

 culture reactions must be gone through before any particular organism 

 isolated is identified as the one or the other ; probably there are saprophytes 

 existing in nature which only differ from them in one or two reactions. 

 In the examination of water, the addition of '2 per cent carbolic acid to 

 the medium inhibits to a certain extent the growth of other bacteria, 

 while the b. typhosus and the b. coli are unaffected. In examining 

 waters, the ordinary plate methods are generally used, but the Conradi- 

 Drigalski or MacConkey media may be employed with advantage. 

 Klein niters a large quantity through a Berkefeld filter and, brushing 

 off the bacteria retained on the porcelain, makes cultures. A much 

 greater concentration of the bacteria is thus obtained. From time to time 

 various substances havebeen used with the object of inhibiting the growth 

 of the b. coli without interfering with that of the b. typhosus. Most 

 of these have not stood the test of experience. Lately caffeine has been 

 used for this end. For use in examining waters the following is the 

 method employed. To 900 c.c. of the suspected water there are added 

 10 grammes nutrose dissolved in 80 c.c. of sterile water, and 5 grammes 

 of caffeine dissolved in sterile distilled water heated to 80 C. and cooled 

 to 55 C. before addition. After mixing the ingredients there is added 

 10 c.c. of '1 per cent crystal violet. The flask is incubated at 37 C. for 

 12 hours and then plates of Conradi-Drigalski medium are inoculated from 

 it. For investigation of faeces, a medium made up as above but with 

 ordinary sterile water may be infected and a similar procedure followed. 

 On the whole there is little to be gained from this attempt to isolate the 

 typhoid bacillus from water in any particular case, and it is much more 

 useful for the bacteriologist to bend his energies towards the obtaining 

 of the indirect evidence of contamination of water by sewage, to the 

 nature of which attention has been called in Chap. IV. 



