290 BACTERIOLOGICAL AND ENZYME CHEMISTRY 



contact bed, that is, a tank filled with large pieces of coke, 

 burnt clay or other material. Such a tank is first filled up 

 with the sewage, when the insoluble and colloidal matters 

 are, for the most part, deposited upon the surfaces of the 

 pieces of material in the tank, and the liquid allowed to run ofi 

 from the bottom of the tank. Air enters the interstices of the 

 medium to replace the hquid, the tank is allowed to remain 

 empty for some hours, and opportunity is afforded for the 

 deposited organic matter to be oxidised ; the tank is then 

 filled again with sewage and the cycle of operations repeated. 

 A tank of this kind may be termed an aerobic tank, and the 

 changes which go on in it are essentially different from those 

 taking place under anaerobic conditions, as described under (6). 

 The process used at Sutton acted well in so far that the 

 hea\aer suspended matter in the sewage was largely removed, 

 and converted in course of time into a nearly odourless 

 residuum. The main drawback to the process was the gradual 

 blockage of the interstices of the medium and the difficulty 

 of cleaning it without complete removal from the tank. This 

 difficulty Dibdin seeks to avoid in his recently introduced 

 slate bed. In this case, instead of the tank being filled with 

 irregular lumps of material, superimposed horizontal layers of 

 slate are made use of, separated by distance pieces about 

 two inches thick. On filling the tank with sewage, the 

 suspended soHds deposit themselves on the slates, and are 

 gradually oxidised in the same manner as in the Sutton 

 process. It is possible to remove the deposit from time to 

 time from the surface of the slates by flushing out, and so to 

 retain the water-holding capacity of the tank undiminished. 

 The writer has had occasion to examine with some care the 

 changes which go on in these slate beds, as they are called. He 

 found that the oxidation of the organic matter, and especially 

 of the fatty constituents, is largely due to masses of nematode 

 worms, with infusoria, etc., and, of course, bacteria. The 

 deposit on the slates, in course of time, assumes a hver-like 



