154 MICRO-ORGANISMS IN WATER 



to the value of sand-filtration as a means of water puri- 

 fication ; but the responsibility which we have seen 

 attaches to this treatment of water cannot be exagge- 

 rated, for whilst when efficiently pursued it forms a most 

 important barrier to the dissemination of disease-germs, 

 the slightest imperfection in its manipulation is a con- 

 stant menace during any epidemic. 



It will be necessary now to turn our attention to 

 those investigations of practical interest which have 

 been carried out on an experimental scale. 



Experimental Sand-Filtration. Fraenkl and Piefke 1 

 studied the question of sand-filtration by taking the 

 Berlin water and passing it through specially con- 

 structed sand filters, which were arranged so as to 

 imitate as nearly as possible in miniature the process of 

 filtration as conducted on the large scale at the water- 

 works. Their object was to trace, if possible, the effect 

 of varying the several conditions under which filtration 

 is carried out upon the number of micro-organisms in 

 the filtered water. 



In the first instance filtered Berlin water and new 

 sand were used. In order to find out whether the 

 organisms contained in the water supplied to the filter 

 were to be found in the filtrate, they purposely intro- 

 duced large numbers of the Bacillus violaceus (see p. 472) 

 into the unfiltered water. This is a non- pathogenic 

 organism, found only very occasionally in water; it 

 produces a violet pigment in the media in which it is 

 cultivated, and hence is readily recognisable amongst 

 other organisms on a gelatine-plate. Two filters were 

 used, one (A) allowing of a rapid (300 mm. in an hour), 

 and the other (B) of a much slower rate of filtration 

 (100 mm. in the hour). It was found that the 

 difference in the rate of filtration produced a marked 



1 Zcitsclirift fiir Hygiene, vol. viii. 1890, p. 1. 



