400 



REPORT — 1878. 



■with the quantity pumped. If this quantity should be large, the bottom 

 of the suction pipe, or, in other words, the apex of the water cone, would 

 have to be carried down until the base of the water cone is enlarged to an 

 area sufficient to supply the requirements of the pump. If that area is 

 partly covered with anything which is impervious, the base of the water 

 cone would extend wider and wider till it reached some source of water, 

 and then, like blotting paper, the exhausted rock would absorb it and 

 pass any excess to the flowing stream which leads to the pump. 



In Liverpool there are several wells within one mile of the docks and 

 river Mersey, yielding daily several million gallons of water. The yield 

 has been continuous for many years. The water has become more 

 brackish each year, some of the wells yield water half as brackish as sea 

 water, but it has always been a mystery where water less salt than sea 

 water percolates so largely into them. 



The expriments just described will, I think, dispel the mystery and 

 give data for calculating the quantity of water freed from salts any given 

 area and depth of similar sandstone rock would filter. 



The next question I wished to decide was — Is the filtration through 

 the sandstone a mechanical or a chemical process ? I will give the answer 

 as follows : — 



One of the cubes of stone which I had used as a filter in the first 

 experiment was allowed to dry in air for a month, and then I poured spring 

 water into the dished part, as I had done with the sea water. Following 

 are the results obtained, and given here in tabular form : — 



Taking sea water at 100 as the standard for comparison, we see that 

 in the first filtrate of 24 fluid ounces there was an increase of 57'77 of 

 the chlorides, and the third filtrate shows that it required 101 fluid ounces 

 of water to reduce the salts which had accumulated in the pores of the 

 stone cube during the filtration of the sea water to the standard of the 

 original sea water. The sixth filtrate shows that 92 fluid ounces ad- 

 ditional of spring water washed out all the remaining chlorides which the 

 cube had taken out of the sea water. The last drops of the sixth filtrate 

 only showed a trace of salts remaining. 



It appears, therefore, that the filtering action is purely mechanical or 

 molecular, and not the result of any chemical action by the rock upon 

 sea water. 



If I may hazard an opinion in a report intended to be free from 

 assumptions, I would suggest, as a probable explanation of the results 

 obtained, that the capillary attraction of the grains of sand of which the 

 rock mass is composed is more powerful for the sodium, calcium, and 

 magnesium chlorides, than it is for the water of solution. These are, there- 



