384 Dr. P. F. Frankland. [June 18, 



faction of the gelatine, whilst a number of those in the unfiltered 

 water produced liquefaction of this medium. 



Brickdust. — The following results were obtained with a filter con- 

 structed similarly to the above, and charged with pulverised red 

 brick, the powder being passed through the before-mentioned sieve. 



Initial efficiency. 



Unfiltered water contained. . 3112 centres per c.c. 

 Filtered „ „ . . 732 „ „ 



After being in action for five weeks, during which time 12' 75 litres 

 of water passed through the filter — 



Unfiltered water contained. . 5937 centres per c.c. 

 Filtered ,, ,, . . 406 „ ,, 



This material, therefore, does not wholly remove the organisms, 

 even when fresh, but it continues, even after five weeks, to remove a 

 considerable proportion of them. 



Coke. — With this material, which was also used in the same state 

 of division, the following results were obtained : — 



Initial efficiency. 



Unfiltered water contained. . 3112 centres per c.c. 

 Filtered „ . . No organisms whatever. 



After being in action for five weeks, when 1325 litres of water had 

 passed through — 



Unfiltered water contained. . 5932 centres per c.c. 

 Filtered ,, „ . . 86 „ ,, 



This substance, therefore, possesses filtering powers which are 

 second only to those of spongy iron ; at the outset the result is equally 

 perfect with both, and even at the close of five weeks' continuous 

 work the coke removes a large proportion of the micro-organisms 

 present in the unfiltered water. Moreover, the rate of filtration 

 through coke was considerably greater. In the case of the coke, as 

 in that of the spongy iron, the organisms found in the filtered water 

 were almost exclusively of one kind, only one colony causing lique- 

 faction, forming an exception. 



Silver Sand. — Owing to the highly ferruginous character of the 

 greensand employed in the previous experiments, and bearing in mind 

 the well-known antiseptic properties of salts of iron, it was deemed 

 advisable to investigate the filtering power of sand free from iron. 

 The silver sand employed for this purpose was digested for some days 

 with hydrochloric acid, the latter then carefully washed out, and the 

 filter sterilised as already described. The following results were 

 obtained with this material : — 



