XXIV. J. M. SMAIL. 



works filters at the rate of 100 millimetres, if based on the 

 calculation above, has but a conditional value. 



The opinion that filtered water containing more than 100 

 germs is not sufficiently cleansed, lias been completely justi- 

 fied by the experience of the Altona Water Works, which 

 is confirmed by that of other works. Of course this state- 

 ment is not to be understood to mean that water containing 

 101 or 105 germs per cubic centimetre is to be rejected 

 without more ado. Each case must be intelligently judged 

 by itself, and the number 100 is merely intended to afford 

 to those called upon to form such judgments a basis founded 

 on experience. 



Special prescriptions as to the cleansing and filling of the 

 filters, the limit to which the sand layer may be allowed to 

 waste, the removal of the first water after the putting in 

 of new sand and after each cleansing, are not necessary if 

 the working of the filters is subjected to regular bacterio- 

 logical control and the water which must according to the 

 result of bacteriological examination be regarded as in- 

 adequately filtered removed. It is the manager's affair to 

 take care that tlie filtered water always fulfils the bacterio- 

 logical demands. The construction and that treatment of 

 the filters which yield the water freest of germs will always 

 be the best. 



Each water works will have to construct its own rules 

 with the help of bacteriology; especially it will have to 

 find out how long its unfiltered water requires to form a 

 good filtering mud layer, how much water must remain 

 unused after the cleaning owing to its containing too many 

 germs, how far the sand layer may be allowed to waste, 

 etc. It is also the mauager's business to ascertain the best 

 remedy, if, as so often happens, too great demands are made 

 on the water works, and regular filtration is thus rendered 

 impracticable. In one case the only remedy will be the 



