182 



ter as quickly as possible, and the oxidation of the fresh sewage in 

 biologic filters in case additional treatment is demanded. 



The first step is accomplished by passing the sewage through the 

 grit chamber, in which the heavier mineral particles are deposited. 

 At the Testing Station, for experimental purposes, the sewage then 

 passes in part to the septic tanks and in part to the settling tank. In 

 the former, anaerobic decomposition of the sewage goes on, eventu- 

 ally resulting in a series of chemical changes which, from the point of 

 view of purity, often render the effluent a great deal worse than the raw 

 sewage entering the tank. The sludge is removed from the settling 

 tanks at frequent intervals while it is still in the early stages of de- 

 composition, the effluent being improved by the process. The prin- 

 ciples of the septic and settling tanks are combined in the Emscher 

 or Imhoff tank, which is a device permitting the escape of a "fresh" 

 effluent, while the suspended matter settles, dropping into a separate 

 sludge digestion chamber. The modified Dortmund tank (a kind of 

 settling or biolytic tank) is of such construction that sulphate re- 

 duction, due to bacterial activity, is typical, and at times as much as 

 40 p.p.m. of hydrogen sulphide have been observed in its effluent. 



The second step consists in biologic treatment or oxidation of the 

 sewage. This is accomplished in the sprinkling filters, which receive 

 the various effluents from the septic and settling tanks. Since these 

 filters are intimately connected with the work a brief description of 

 them will be given. Sprinkling filters are not built primarily as a 

 device for removing suspended matter, but as a means of oxidizing 

 and mineralizing the organic matter delivered to them. Each sprin- 

 kling filter consists essentially of a bed of crushed limestone, 4^ to 

 10 feet deep, and a central top-surface intermittent spray, which con- 

 stitutes the influent. The size of the stone, the depth of the bed, and 

 the period of the spray is different in each filter. In sprinkling filter 

 No. 4, the flow-period is about 60 seconds with an intermission of 

 about 30 seconds. The daily flow is approximately 10,000 gallons. 

 Sprinkling filters function to a considerable extent in the catching of 

 suspended matter, and as a consequence sludge accumulates on the 

 stones. The chemical nature of this sludge depends of course upon 

 the character of the influent. 



According to the observations of Dr. Lederer and others con- 

 nected with the testing station, Lumbricilhts rutilus has a seasonal 

 distribution. The worms apparently disappear completely at the ap- 

 proach of winter (November or early December) but in March or 

 April they begin to appear in all of the sprinkling filters and their 

 effluents. In a short time they become extremely abundant in the 



