I 



CHEiMICAL CHANGES 125 



diminishes after 3 feet, and that there is no nitrification below 

 6 feet. The best results from bacterial filter-beds are obtained 

 with a depth of 3 to ^j feet, and an alteration to 5 feet is 

 generally accompanied by a reduction in the quality of the 

 effluent. Here I may draw attention to a fact that awaits 

 explanation. In nearly all published analyses the chloride in 

 the effluent is slightly lower than that in the corresponding 

 sewage. Miintz pointed out that in nitrification bromides and 

 iodides were oxidized to bromates and iodates. Chili saltpetre, 

 nitrate of soda which has been produced by natural nitrifica- 

 tion, often contains a small percentage of perchlorates. Dr. 

 Tidy, some twenty years ago, found a loss of chloride in waters 

 running over aerating wooden shelves, and suggested that it 

 might be due to the formation of chlorates. I have not yet 

 been able to find them in effluents. 



Both in nitrification and in the fermentations producing 

 organic acids the alkalinity is diminished, and the extent to which 

 this occurs gives some measure of the degree of the change ^ 

 (p. 32). Carbonates of lime and magnesia and other earthy 

 salts may be dissolved, the former with liberation of CO2 ; in 

 both cases the mineral matters in solution will be increased. 

 Dibdin's averages of the " fixed dissolved solids" at Exeter in 

 1897 were, in parts per 100,000 : Raw sewage, 20*0 ; tank 

 effluent, 23*0 ; filtrate, 29*0. Such reactions also have a ten- 

 dency to disintegrate some filtering materials (see Chapter X.). 



At this stage an abundance of carbonic acid is formed by 

 fermentations due to other classes of bacteria. 1 have found 

 in several bacterial filters intended to be aerating and final 

 such a large quantity of carbonic acid as must seriously retard 

 their nitrifying action, the result being a deficiency of nitrates 

 in the effluent. Especially is this the case where the final beds 

 are made by a process often recommended for economy — that 

 of simply digging out the clay to form a pit about 3 feet deep, 

 and filling it up with the same clay after burning. 



Boulanger and MassoP have confirmed the fact that nitrous 

 and nitric organisms (i) are each hindered when the products 

 of their own action accumulate ; (2) tend to hinder one another, 

 and therefore should be kept as far as possible in separate 

 areas. See also H. Chick on Nitrification, Proc. R. Soc, 1906, 

 p. 241. 



Denitrification has been largely investigated for the reason 



1 Massachusetts Report, 1905, p. 227. '^ Ann. Inst. Pasteur ^ 1903, 492. 



