Report on the Bacteriology of Water. 463 



Tims, in the case of the B. coll communis again, there was the same 

 disappearance in five days of the bacilli in the water kept at 19 C., 

 the great diminution in numbers in the same time in the water kept 

 at 6 8 C., followed by complete disappearance of the bacilli in this 

 water also by the twelfth day. Similar attempts made by the addi- 

 tion of sterile broth to resuscitate the bacilli in these waters also 

 proved unavailing. 



These results, showing that the typhoid and coli bacilli were more 

 rapidly destroyed in the porcelain-filtered than in the unsterilised, 

 and far more rapidly than in the steam-sterilised, Thames water, 

 were so surprising that it was necessary to banish every suspicion 

 of some accidental disturbing cause having arisen in these experi- 

 ments. 



The most obvious suggestion was that the filter itself might have 

 introduced some antiseptic substance into the water. This was, 

 however, highly improbable, as the filter in question had only been 

 previously used for the similar sterilisation of Thames and Loch 

 Katrine waters. In order to abolish this objection, however, the 

 porcelain cylinder was thoroughly scrubbed externally with a tooth- 

 brush, and then upwards of 30 litres of distilled water passed 

 through it. The filter was then stefm sterilised and employed for 

 the filtration of some more of the same Thames water, which was 

 infected with typhoid and coli as below. Thus 



(<z.) Typhoid Bacillus. 20 needle-loops were taken from an agar 

 cultivation of the typhoid bacillus of nine days age and introduced 

 into 50 c.c. of steam-sterilised water ; after thoroughly shaking, 

 3 c.c. of this water-attenuation were added to 750 c.c. of the por- 

 celain-filtered Thames water. 



(6.) Bacillus coli communis. The infection was made in exactly the 

 same way in every detail, the agar culture being also of the same 



'age- 



The waters thus infected with typhoid and coli respectively were 

 subdivided into smaller flasks, some of which were placed as usual in 

 the incubator at 19 C. and others in the refrigerator at 6 8 C. 

 The following results were obtained on examination: 



