222 SEWAGE AND ITS PaRIFICATION 



In this case we have actually a reversal for a time of the 

 functions of the two beds, and a violation of the law that *' the 

 bacterial changes should be carried out with regularity and in 

 natural sequence." 



A great divergency was noticed in the successive individual 

 samples, taken at fifteen minutes' interval, and can be ac- 

 counted for by the interference of the filling material with the 

 free mixing of the sewage, so that zones and channels are 

 formed through which the liquid flows at varying rates. The 

 effluent issuing at successive intervals of time comes from 

 different layers and parts of the beds, and really represents 

 sewages of different hours or even days, as proved by the 

 individual chlorine figures. 



There is a considerable loss of nitrogen in contact-beds 

 (see pp. ii8 and 126 in Chapter V.). Letts and Lorrain 

 Smith investigated this at Belfast,^ and their chief results 

 are : 



1. They confirmed previous observations that a main cause, 

 reaction between nitrates, nitrites, ammonia, and organic 

 matter with production of free nitrogen and CO2, was due to 

 vital processes of organisms and not to enzymic or chemical 

 action, and that the common B. colt was one of the species 

 taking part in the change.^ 



2. Green algai sometimes flourished on the bacteria beds, 

 and absorbed large quantities of ammonia and nitrates ; their 

 tissues contained a larger proportion of nitrogen, and less 

 mineral matter, than when they were growing in ordinary 

 waters.^ [Compare the vigorous growth of grass and weeds 

 that generally forms, if allowed, on inland filter-beds. In 

 America, sand filtration beds have been regularly planted and 

 cropped.] 



3. In the brick beds at Belfast, " 12 to 20 per cent, of the 

 nitrogen disappearing can be recovered as free N dissolved in 

 the effluents," and a further portion escapes as nitrogen gas or 

 oxides of nitrogen. 



4. An uncertain amount passes into the tissues of animalculas, 

 worms and insects (see Chapter IV., p. yy). The bacterial 

 jelly, or zoogloea, encrusting the filtering material also contains 

 much nitrogen. 



Phelps and Farrell have demonstrated experimentally the 



1 Belfast City Reports, igoi and 1904; Chetn. News, No. 2184, 1901. 



2 See Pakes and Jollyman's Papers, Trans. Chem. Soc, 1901, pp. 322, 386, 459. 



3 Belfast Report, 1904, p. 46. 



