90 REPORTS ON THE S1ATE OF SCIENCE.—1918. 
The evidence available with respect to the mechanism of the 
removal fiom sewage of organic matter in pseudo-solution may be 
summarised thus :— 
(i) The experiments of Stoddart designed to show its direct 
oxidation by bacterial agency cannot be considered 
conclusive. On the other hand it has not been demon- 
strated that under the favourable conditions present in an 
efficient filter, in which the sewage passes in thin films 
over a network of media covered with an active bacterial 
slime, the mean time of passage is insufficient to allow of 
direct biolytical oxidation of some portions of the organic 
matter in pseudo-solution. 
(ii) While it is accepted that sewage colloids may be precipitated 
or coagulated by intimate surface contact, this ‘de-solution’ 
theory can scarcely represent the whole phenomena of 
colloid removal otherwise such action would continue 
in the absence of micro-organisms. 
(iii) The conditions met with in a mature and efficient filter are 
such as to render it most probable that absorption phe- 
nomena play an important part in the removal of matters 
from pseudo-solution. 
In considering the final fate of the matters removed from pseudo- 
solution during the purification process it may be said that, inde- 
pendent of what may occur with experimental filters inoculated with 
vigorous growths of micro-organisms, and operated under ideal 
conditions, Stoddart’s view that none of the solid matter retained in 
sewage filters is derived from organic matters originally in pseudo- 
solution is not consistent with what obtains in actual practice. 
On the other hand the contention of Travis that the retained 
colloids undergo very little change as the result of biolytic action 
was formulated on entirely inconclusive evidence. 
In this connection mention may be made of Dunbar’s experiment 
in which a solution of albumen containing nitrogen equivalent to 
the organic nitrogen content of an average sewage was applied toa 
mature filter, with the result that almost quantitative oxidation of 
the sulphur was obtained, while 58 per cent. of the total nitrogen 
appeared in the filtrate, 10 per cent. as ammonia, 20 per cent. as 
nitrate and 25 per cent. as organic nitrogen. Thus 42 per cent. of 
the nitrogen disappeared either in the gaseous form or accumulated 
in the ‘humus.’ A considerable part of the carbon of the albumen 
also disappeared as gaseous carbonic acid, but a part also went to 
form the ‘humus’ which accumulated in the filters. 
The Massachusetts intermittent sand filters have demonstrated that 
the amount of retained nitrogenous matter after 18 years’ continuous 
operation with ‘regulation station’ sewage containing its original 
suspended solids*only amounts to from 4 to 5 per cent. of the total 
nitrogen in the sewage applied. Travis estimates this to be equal to 
about 20 per cent. of the original organic nitrogen. 
Recently, Clark in reporting on these filters after they had been 
at work for 28 years, states that for some years only as much sewage 
has been applied as can be purified without increasing the amount of 
organic matter stored in the filters. 
