BACTERIAL PURIFICATION 277 



more economically conducted in a tank which is continuously 

 full, than in a filter bed constructed for aeration. Nitrification 

 is usually most energetic in liquids in which the organic matters, 

 especially carbohydrates,^ are a minimum, and therefore a 

 denitrification change effected during the anaerobic preliminary 

 stage, by reducing the oxygen-consumed figure to a greater 

 extent than would be the case if the change were only due to 

 hydrolysis, yields an effluent which contains its nitrogen in the 

 most available form for the changes of which nitrification is the 

 final result. Thus, for example, a hydrolysed effluent with, say 

 an oxygen-consumed figure of 3, and unoxidized nitrogen 

 10 parts, on passing through the filters often yields not more 

 than 3 parts of nitric N ; if, however, a portion of this liquid be 

 returned to the tank, it will so reduce the O consumed figures, 

 as to allow the nitrification to approximate to the theoretical 

 amount. 



The Report of the experts finally recommended that the 

 sewage be submitted to an efficient process of screening (or to 

 roughing tanks), then passed through open tanks with sub- 

 merged walls and floating scum-boards, and afterwards over 

 60 acres of double-contact beds 3*33 ft. deep, with four fillings 

 per 24 hours. As the dry-weather flow was 30 million gallons, 

 the filters dealt with half a million gallons per acre, allowing 

 one day per week rest. The open tanks were to hold 15 million 

 gallons, and would therefore change their fluid contents in dry 

 weather every twelve hours.^ 



As to storm- water it was found that there was even a 

 greater amount of oxidizable matter in the first flush than in 

 ordinary sewage. They advise that the storm-flow should be 

 dealt with in the same system, and the excess, after passing 

 through roughing tanks, should be taken to specially-prepared 

 bacteria beds of at least 25 acres area. No decrease in efficiency 

 was noticed after the storm had passed.^ 



The Local Government Board decided that a larger area of 

 land must be purchased, " over or through which the sewage 

 after it had left the bacteria beds should pass," also (Oct. 4, 

 1899) '' that not less than 92 acres of filter beds shall be pro- 

 vided for the treatment of sewage by double contact ; and that 

 the filter beds be worked in cycles with the usual provisions as 



1 R. Warington, J. Chem, Soc, May, 1900. 

 - Report Manchester Rivers Committee^ Jan. 22, 1900. 



^ A critical analysis of the Manchester experiments, by Boyce and McGowan, 

 is given in the 2nd Report of the R. Sewage Commission, 1902. 



