SEWAGE OUTFALLS AND DISCHARGE 313 



charging into the sewage sewers to separate the foul street 

 washings from the later discharges of heavy rainfalls. 



Storm-water passing rapidly off the land carries with it 

 disease germs, as is shown by the repeated occurrence of epi- 

 demics when a sudden storm succeeds a period of drought. 

 But the liquid is ordinarily supplied with abundance of the 

 liquefying and oxidizing bacteria, which if it be allowed to 

 subside in auxiliary reservoirs will effect its purification rapidly, 

 aided by the oxygen derived from the air, and by the nitrites 

 and nitrates that rain-water always contains. The sand, chalk, 

 or especially the clay, may be a long time in subsiding, but 

 when deposited will leave the water comparatively pure, and 

 fit for flushing sewers, watering roads, or for supplying the 

 deficiency in rivers during dry seasons. 



Whatever system be adopted the raw storm-water of populous 

 districts should never be allowed to pass in large volumes at the 

 beginning of a storm directly into a stream. The general con- 

 sensus of opinion is that if the first foul storm-water be treated 

 as ordinary sewage, the subsequent rain-flow becomes so dilute 

 that it can be discharged, with only a slight treatment, into a 

 river. The Manchester experts placed a limit of time of two 

 hours after the commencement of the storm. Many towns adopt 

 a volume limit. Thus, Mr. A. M. Fowler, at, Stockport, made 

 provision for an escape after eight times the dry-weather flow ; 

 other places in Lancashire and Yorkshire allow 6 or even 5 to i. 

 By the Leicester Extension Act, 1891, the overflow culvert came 

 into action when the rainfall increased the dry-weather flow of 

 35 gallons per head to 60 gallons, but in this case the overflow 

 passes into the river Soar, which has a flow during dry weather 

 of only about 6 or 7 million gallons per day, so that the storm- 

 water is actually useful for flushing the river bed. 



I have found from analyses — 



1. That, after the first flush, the chlorine content varies with 

 the rainfall. 



2. That with low chlorine and high rainfall, higher nitrifica- 

 tion is obtained. 



3. That, as might be expected, the later diluted sewage 

 comes within the usual standards of permissible impurity, 

 therefore could not, under them, be excluded from streams. 



As I remarked at the Manchester Congress of the Royal 

 Sanitary Institute,^ purification to a bacterial standard of the 



^ Journ. R, San. Inst., vol. xxiii., part iv., 1903, p. 617. 



