WATER FILTRATION. XXI. 



I have only quoted some of the rules, the full text can 

 obtained from "Purification of Sewage and Water," by 

 Mr. Dibdin, F.I.C., etc. 



To shew the importance of regulations in connection with 

 this question, and in view of the prevalence of cholera on 

 the Continent and possible introduction into Great Britain, 

 and importance of efficient filtration of the metropolitan 

 water supply, he requested the particular attention of the 

 directors of water companies to the following points, viz. : — 



" 1. That every effort should be made to maintain the layer 

 of sand in each filter at the greatest practicable thickness. 



u 2. That the rate of filtration should be as slow as possible 

 consistently with the supply of the required quantity of 

 water. 



"3. That the sand removed from the surface of the filters 

 with the deposit should, if it was to be replaced in the filter, 

 be completely freed from all taint of organic matter." 



In a paper read before the Institute of Civil Engineers, 

 November 21th, 1896, Dr. Percy Frankland observed, in re- 

 ference to a former paper read by him in 1886, as follows : — 



"The principles then enumerated were as follows: — (1) To give 

 the maximum period of storage for unfiltered water. (2) To filter 

 at a minimum rate. (3) To filter through a maximum depth of 

 fine sand. (4) To frequently renew the filtering materials. In 

 reviewing the numerous investigations which have since been 

 carried out on this subject, it will be shewn that the only one of 

 these principles which require any modification in the light of 

 more recent researches is the last. Further researches appear to 

 shew that the power of arresting bacteria possessed by a filter bed 

 suffers no diminution with age, so that frequent scraping is not 

 necessary for the maintenance of the efficiency of the individual 

 bed. It is, however, quite possible that it may be of advantage 

 for the efficiency of the filtration plant taken as a whole. Thus, 

 by running a filter bed over a long period of time without cleansing, 



