EXAMINATION OF WATER. 429 



that now develops is placed, by means of a drawn-out tube 

 pipette (Fig. 61), into sterile milk tubes. The inoculated 

 milk tubes are then placed at 39 for 1-3 days. As a 

 trol of the sterility of the milk a number of the uninocula- 

 ted tubes should likewise be placed in the incubator. 

 Moreover, it is advisable to inoculate a like number of 

 milk tubes with a known typhoid bacillus in order to elim- 

 inate a possible error (p. 163), which might arise if spores 

 of anaerobes were present. The milk tubes that coagu- 

 late can be discarded at once, together with ihe bouillon 

 cultures from which they were inoculated. The milk tubes 

 that do not coagulate may, or may not, contain the Eberth 

 bacillus. The bouillon cultures from which these tubes 

 were inoculated can now be used for the supplementary 

 tests. 



The procedure as outlined is preferable to direct inocu- 

 lation into milk fr*om the suspected colonies. The trans- 

 plantation in the latter case may sometimes fail to de- 

 velop or does so very slowly. When bouillon cultures are 

 grown, the inoculations can be made with liberal amounts of 

 the organism and the original material is saved for subse- 

 quent tests. 



At times it may be more advantageous to make a sur- 

 face touch colony on Stoddart's medium contained in a 

 large flask or Esmarch dish. The transparent border 

 growth can then be plated on gelatin or on Eisner's me- 

 dium, and the suspicious colonies can be tested as in 

 the manner indicated above. The addition of a drop of a 

 bouillon culture to a liter of tap-water can be detected 

 quite readily in this way. The methods mentioned above 

 are also resorted to when isolating the Eberth bacillus 

 from the urine or f eces of typhoid patients. 



Cholera vibrio. This organism has been repeatedly 

 found in water during times of cholera epidemics. As in 

 the case of the Eberth bacillus, the detection of the comma 



