236 SEWAGE AND ITS PURIFICATION 



ordinary coarse bed. The cost of slate beds is from 6s. to los. 

 per square yard. Cost per '^ efficiency yard" 3s. to 5s. A bed 

 at Devizes has been working eighteen months, with an abnor- 

 mally heavy sewage, unscreened and unsettled, followed by 

 a fine bed. Although the raw sewage frequently contains 

 200 grains of suspended matter per gallon, and the fillings are 

 two or three times daily, the capacity has never been reduced 

 below 50 per cent. ; then by opening the valve to full bore it 

 was flushed to 64 per cent. More thorough flushing increased 

 it to the original capacity, 82 per cent, of the cubic content. 

 Thus he finds no necessity for removing or renewal of contact 

 material, while the size of the beds need be only half that 

 of coke coarse beds. The deposit on the slates contained 

 3,000,000,000 bacteria per c.c. 



The Ames-Crosta Co. make a special form of tiles for floors 

 of filters, which are easily set, and allow free and uniform 

 passage of liquid and air. 



In the northern portion of the United States, where the 

 severe winter weather interferes seriously with most of the 

 distributing devices, experience at Madison, Wisconsin, and 

 other places has shown that success may follow the application 

 of sewage to filters by means of lines of perforated tile pipes 

 laid close together, and covered to protect them from frost. 



We revert to the processes depending mainly on strong 

 aeration, of which the chief are Lowcock's, Waring's, and 

 Ducat's. In Chapter V., p. 121, we have given a table of the 

 volumes of air required to oxidize the nitrogen of organic 

 matter : a further quantity would be demanded by the car- 

 bonaceous matters, measured approximately by the " oxygen- 

 consumed " figure. It has been shown how in an effluent that 

 has properly passed through all the stages, the residual organic 

 carbon can be disposed of by the nitrates, in presence of the 

 appropriate organism ; but that for an imperfectly hydrolysed 

 effluent, and still more for a raw sewage, a large volume of air 

 is required, and the action is apt to be slow, irregular, and 

 incomplete. This is well seen from experiments^ in which 

 Dr. Fowler exposed a chemically precipitated effluent (lime and 

 copperas) to the air in thin layers, protected from dust, for 

 various periods and under different conditions. In no case 

 was sufficient oxidation effected in twenty-four hours to render 

 the effluent subsequently non-putrefactive. Even after seventy- 



^ Manchester City Surveyor's Report for 1897. 



