BACTERIA IN WATER 75 



is necessary. Not only may the age of the filter act preju- 

 dicially, but the extra pressure required will tend to force 

 through it bacteria which ought to have remained in the filter. 

 In 1893, Koch brought out his monograph upon Water 

 Filtration and Cholera, and his work had a deservedly great 

 influence upon the whole question. He shows how the 

 careful filtration of water supplied to Altona from the Elbe 

 saved the town from the epidemic of cholera which came 

 upon Hamburg as a result of drinking unfiltered water, al- 

 though Altona is situated several miles below Hamburg, and 

 its drinking water is taken from the river after it has received 

 the sewage of Hamburg. Now, from his experience of 

 water filtration, Koch arrived at several important conclu- 

 sions. In the first place, he maintained that ih^ portion of 

 the filter-bed which really removed -micro-organisms effectively 

 was the slimy organic layer upon the surface. This layer is 

 produced by a deposit from the still unpurified water lying 

 immediately above it. The most vital part of the filter-bed 

 is this organic layer, which, after formation, should not be 

 disturbed until it requires removal owing to its imperme- 

 ability. A filter-bed, as is well known, consists of say three 

 feet of sand and one foot of coarse gravel. The water to 

 be filtered is collected into large reservoirs, where subsi- 

 dence by gravitation occurs. Thence it is led by suitable 

 channels to the surface of the filter-bed. Having passed 

 through the three or four feet of the bed, it is collected in a 

 storage reservoir and awaits distribution. The action of the 

 whole process is both mechanical and chemical. Mechanic- 

 ally by subsidence, much suspended matter is left behind in 

 the reservoir. Again, mechanically, much of that which 

 remained suspended in the water when it reached the filter- 

 bed is waylaid in the substance of the sand and gravel of the 

 filter-bed. Chemically also the action is twofold. Oxidation 

 of the organic matter occurs to some extent as the water 

 passes through the sand. Until recently this chemical action 



