158 MICRO-ORGANISMS IN WATER 



minimum (25 mm. per hour), some, although very few, 

 colonies of Bacillus violaceus were obtained from the 

 filtrate. Again, in similar experiments in which cholera 

 and typhoid bacilli were purposely added to the un- 

 filtered water, these were also met with, in greatly 

 diminished numbers of course, in the filtrate ; and it 

 is worthy of note that the cholera appeared to be more 

 completely retained by the filter than the typhoid bacilli. 



Piefke, 1 who is the resident engineer at the Berlin 

 waterworks, has made some instructive investigations 

 as to the manner in which the coating of slime on the 

 filter beds affects the results of filtration. 



A kilogram of sand taken from one of the Berlin 

 filter beds was found when examined from the surface 

 to contain 5,028 millions of bacteria, about 2 cm. deeper 

 734 millions, 10 cm. 190 millions, 20 cm. 150 millions, 

 and 30 cm. 92 millions ; whilst the layer of fine flints 

 upon which the sand rests contained 68 millions. These 

 examinations show that the upper layers of slimy sand 

 afford a very important barrier to the passage of micro- 

 organisms. When a filter starts working the micro- 

 organisms present in the filtering material and intro- 

 duced in the water are partially washed through ; those 

 which remain behind become attached to the sand 

 particles, multiply and produce the well-known slimy 

 coat. 



These multiplying centres of bacteria, which at first 

 are more especially found in the upper layers of the 

 filter, serve as points of attraction and adhesion for 

 those bacteria which have not been retained by the 

 slime. This coating of slime gradually then spreads to 

 the lower layers of the sand, which becomes slimy to the 

 touch, losing all feeling of sharpness, throughout the 



1 Die Principien der Eeimuassergewinnung vermittelst Filtration. 

 Berlin, 1887. 



