214 SEWAGE AND ITS PURIFICATION 



noted that it is earlier stated in the report that " the sewage 

 had been roughly screened before reaching the coke-beds, and 

 was free from larger matter usually described as ' filth,' and 

 from coarse sand and heavy mineral road detritus . . .," so 

 that, as I have always insisted, the additional sedimentation 

 would mean a further evasion of complete bacterial treatment, 

 and a production of a further amount of supplementary sludge. 



A strong confirmation of the suggested origin of the loss of 

 capacity is found in the remark that " the ash in the coke has 

 been reduced in amount by about 25 per cent, during its ex- 

 posure to sewage in the coke-beds," cellulose being nearly 

 ashless. 



These results confirm the view which has been frequently 

 urged — namely, that these non-animal substances cannot be 

 successfully destroyed without anaerobic action (see Chapter V.), 

 by which they are dissolved with production of gas. They are 

 the great difficulty in all processes where the first or hydrolytic 

 change is not properly specialized. The degree to which the 

 nitrogenous matter is dealt with cannot be traced from the 

 report, as only the "oxygen-absorbed" figures are given, and 

 it is obvious that if the non-nitrogenous matter is arrested by 

 the filter-bed, the improvement in the effluent as measured by 

 the oxygen-consumed figure must in part be attributed to this 

 cause, at the expense of clogging or diminution of capacity. 



I have already remarked that the first stage requires no 

 oxygen, and is actually hindered by it, the second requires 

 some, while the third demands a very large and rapid supply. 

 In place of providing three separate areas in which these con- 

 ditions are carefully and continuously observed, as we should 

 in the culture of plants which required different amounts of 

 water, heat, or manures, it is attempted to alternate them in 

 two receptacles by causing the air in each to be cut off and 

 supplied intermittently, and the sewage to be either stagnant, 

 or run in and out with a rush, with the result that the bacteria 

 are periodically disturbed, and neither class of organisms can 

 work under their normal vital conditions. My own analyses 

 and those of others have proved that under the intermittent 

 system, first adopted from the laboratory experiments of Sir E. 

 Frankland in 1870, the effluents, although the average results 

 show a great improvement, yet manifest such fluctuations in 

 character, tending to be periodic, as show that the quiet and 

 regular working of the bacteria suffers avoidable interruption 



