216 



Transactions of the Society. 



know which sample they were taking — that one was bright and 

 sparkling in taste, the other insipid, though more frequently than 

 not they hit upon the treated water as being the better sample 

 of the two. I hold, of course, that it is no better, and that 

 when the process is carried out properly no difference can 

 be discerned. In the earlier experiments plates and broth cul- 

 tures had to be prepared in order that the presence or absence 

 of bacteria in the water, both before and after treatment, 

 could be determined, the bacterioscopic examinations being 

 always supplemented by a chemical test for the presence or 

 absence of available chlorine, and it was noted that whenever the 

 chemical test gave an indication of the presence of the slightest 

 excess of chlorine at the end of twenty minutes, the B. c. c. 

 was invariably absent even from such large quantities as five or 

 six samples of 500 c.c. each. It was manifest, therefore, that if, 

 after allowing the chloride of lime solution to act for a period of 

 twenty minutes, there still remained a slight excess of chlorine 

 demonstrable by the iodide of potassium and starch test, the water 

 must be " sterile." These experiments were repeated time after 

 time, always with the same satisfactory results. 



Fig. 23. — Eesult of Bacterioscopic examination of Cambridge U. and T. W. W. 

 Co.'s Fulbourn water before and after treatment with 1 part Chlorine to 

 7 millions of water. 



Treated water thus, " all negative." 



Since these earlier observations more exact experiments have 

 been carried out. In these the action of the hypochlorous acid 

 has been cut short by adding sterile thio-sulphate of soda to the 



