. II. SALT WATER MADE FRESH BY FILTRATION. 43 



or less, completely deprived of salt. I have confirmed this 

 experiment by using a tube of about 8 metres [about 26 feet] 

 long, filled with sand, and I have found that the density of 

 the liquid, introduced by the upper aperture of the tube, was 

 to that of the liquid escaping from the other end, as 1 : 0-91. 

 But it is necessary to state that this difference of density 

 was not always maintained ; after a certain time, the saline 

 solution is as dense at its exit from the tube as at its en- 

 trance. This proves that the decomposition of the -saline 

 solution takes place in the first action of contact between it 

 and the particles of sand.* 



I have obtained an inverse result by employing a solution 

 of carbonate of soda, which I caused to pass through a tube 

 3 metres [nearly 10 feet] long, filled with sand. The den- 

 sity of the liquid at its exit was to that at its entrance as 

 1-005:1. 



The phenomena we have just referred to are very impor- 

 tant on account of the applications that can be made of 

 them to some of the functions of living beings which are 

 not completely explicable by the mere action of capillarity 

 and molecular attraction. 



* The facts here mentioned respecting the alteration in the density of 

 a saline solution by filtration are very interesting. They are susceptible 

 of numerous applications in geology. For example, the marine origin of 

 fresh- water springs has hitherto been deemed impossible, " because," as a 

 recent writer observes, " sea water cannot be freed from its salt by filtra. 

 tion." Matteucci's experiments demonstrate the possibility of this ori- 

 gin.J. P. 



