220 SEWAGE AND ITS PURIFICATION 



was 102 gallons per square yard per day, while the fine filter 

 (83*3 square yards) dealt with the coarse filtrate at the rate of 

 157 gallons per square yard per day, or approximately, for the 

 double filtration, 10 acres for 3,000,000 gallons of screened 

 sewage. 



At that time the distribution of the liquid over the beds was 

 of the simplest type, causing an irregularity of contact between 

 the material and the liquid which was revealed in the analysis. 



The suspended solids in passing the coarse bed fell from 61 

 to 18, the difference of 43 being retained. Their resolution, as 

 I have pointed out, is mainly an anaerobic process, actually 

 antagonistic to the oxidizing and nitrifying changes which are 

 intended to occur in the fine bed. It is proved, however, by 

 the increase of combined nitrogen in solution from 3*4 to 4*1 

 parts, that the fine bed had in this case to supplement the 

 coarse bed in dissolving nitrogenous solids. 



The general lowness of the free and albuminoid ammonias 

 with high organic nitrogens is probably explained by the fact 

 that the effluents were analysed in such a fresh state that the 

 nitrogen was mainly present as urea, since this compound does 

 not readily yield its nitrogen by distillation with alkali or per- 

 manganate, but is completely changed into ammonia by the 

 Kjeldahl process, hence would appear as organic N. It is well 

 known that before urea can be nitrified it must be hydrolysed 

 into ammonia : the first stage should be effected in the coarse 

 bed, the second in the fine. 



On the other hand, during the thirteen hours of rest and 

 aeration that had elapsed before the first samples were taken, 

 the coarse bed had temporarily assumed a nitrifying function, 

 as shown by the very considerable amounts of nitric nitrogen 

 found, with a lower quantity of nitrite, and only a slight re- 

 duction of the oxygen consumed. Later in the day, when the 

 rest periods are shorter, all this nitrate disappears, with a heavy 

 fall in the total nitrogen, and a considerable lessening of the 

 oxygen consumed. There is little doubt that this is explained 

 by a Gayon and Dupetit reaction, by which nitrates and organic 

 nitrogenous matter decompose one another, the oxygen of the 

 nitrate burning up the carbon, and nitrogen or oxides of nitro- 

 gen being evolved as gas. Possibly the disturbance occasioned 

 by the formation of this gas accounts for the extraordinary 

 variations in individual samples, and for the high suspended 

 matter occasionally met with. 



