BACTERIAL PURIFICATION 231 



material more than another, so far as the chemical results were . 

 concerned. . . . Observations were made as to the bacteria in 

 the respective effluents, but no specific advantage seemed to be 

 shown by any one material in this respect." 



As in the Massachusetts, Barking, and Berlin experiments, 

 the sacrifice of capacity and of output on substituting a less 

 porous material, like sand, was not accompanied, in the case of 

 sewage, by an equivalent advantage in purification. 



2. That no nitrate or nitrite was produced in any case in the 

 first beds is a strong indication that they were acting hydro- 

 lytically and anaerobically, their function being that of an 

 " open septic tank." 



3. The double filtration, or result of the second bed, cannot 

 be exactly followed, as the nitrites are not separately given, but 

 it corresponds mainly with the second stage of partial oxidation. 



4. In the triple contact, the coke breeze has shown a higher 

 power of nitrification, as noticed by other observers, owing, 

 undoubtedly, to its greater porosity. 



It is also important to note that these filters prove that the 

 organic carbon is more easily oxidized than the organic nitrogen, 

 and confirm the criticism on the use of coal (p. 227). 



To study the influence on nitrification, in October, 1899, I 

 examined six tray filters after running about three months with 

 a hydrolysed sewage. D, E, and F, had an area of 100 square 

 feet each ; A, B, and C, were one-third the area, and had 

 become much clogged. D was most freely exposed to the air. 

 The filtrates gave on successive days the averages in parts per 

 100,000 on p. 232. 



Denitrification with loss of nitrogen is here shown by those 

 filters which are not in proper order. The superiority of a 

 graded filter (F) is also evident, while coal has exhibited the 

 peculiarity that has been noticed in other cases, of encouraging 

 the production of nitrites. Filter F shows the extraordinary 

 nitration of a strong sewage, resulting in an excellent effluent ; 

 in these cases a gain of total nitrogen, presumably from the air, 

 has often been observed.^ The great variation produced by the 

 ventilation and aeration of similar filters is also seen in com- 

 paring B, D, and E. 



With reference to porous materials the conclusions of the 

 Manchester Report'-^ agree with previous experience in finding — 



^ Some species of bacilli can assimilate free N, as in the agricultural prepara- 

 tion " nitragin," 



- Baldwin Latham, P. Frankland, and W. H. Perkin, October, 1899. 



