XXII. J. M. SMAIL. 



its yield becomes diminished, and as a certain volume of water 

 must be supplied daily this may necessitate filtering through other 

 beds at a disadvantageous high rate, and so obtaining a nitrate 

 from the entire works inferior in bacterial quality to that which 

 would result if scraping were practised more frequently and a 

 more uniform rate of nitration employed with all the beds " 



Dr. Koch, in dealing with water filtration in relation to 

 cholera, says "The question now is what lessons we have 

 to learn fr;)m them for the future?" The numerous and 

 thorough investigations which have been made for some 

 years past at Berlin and Altona Water Works with regard 

 to the process of filtration and the bacterial condition of 

 the water before and after filtration, has led to the convic- 

 tion that filtration rate of 100 millimetres, or 3*97 inches 

 an hour, affords a sufficient guarantee for the satisfactory 

 working of a filter bed. Further experience of those water 

 works, however, has shewn that we do not gain very much 

 by simply making this demand. For, with the now existing 

 arrangements, most water works will not be able to fulfil it, 

 and in point of fact they do not fulfil it. The principle 

 must he adhered to, nevertheless, that in future a filtration 

 rate of 100 millimetres must be the first condition; but we 

 in nst formulate our demand in more precise terms, and so 

 far supplement that the purpose aimed at may be attained 

 with certainity. This is effected by the somewhat extended 

 demands. 



1. The filtration rate of 100 millimetres, or 3*937 inches, 

 must not be exceeded. In order to render this possible, 

 each filter must be provided with a contrivance by means 

 of which also it may be ascertained at any moment whether 

 this rate is observed or not. 



2. Each filter basin, so long as it is at work, must be 

 bacteriologically examined once a day. It must therefore 



