156 TRANSACTIONS OF THE CANADIAN INSTITUTE. [VoL. I. 



Duclaux has stated that of the agencies which render water 

 sterile, " the first, the oldest known and without doubt the most 

 potent is the capillary action of the soil. Filtration practically 

 retains in the capillary interstices the matters in suspension in 

 the water, and with them the germs of microbes. It is a fact well 

 demonstrated on which I shall only insist in order to attempt to indicate 

 slightly what one cjiUs capillary filtration .... It is only proper 

 to remark at the outset that the capillary character of the channels in 

 which circulates the water of rain, has only the effect of augmenting the 

 extent or surface or the volume of water which leaves them, that is to say 

 of multiplying the chances which a solid particle in suspension in water 

 can have of encountering a portion of wall on which it fixes itself, drawn 

 by a force analogous to that which fixes coloring matter in a tissue placed 

 in a color solution. The efifect would be the same if the chances of con- 

 tact were found to be increased by any other cause. It could also 

 happen and sometimes does happen, moreover, that a long repose causes 

 to adhere to the walls of a vase the particles held in suspension in the 

 liquid it contains. It can happen, and without doubt does happen, that 

 a slow filtration through a great length of non-capillary spaces aids and 

 even somewhat largely produces the same result as through capillary 

 spaces shorter and narrower." 



We thus see that the rate at which water passes downward through 

 the soil which as Hofmann found at Leipsic in some virgin soil, was at the 

 rate of 6"2 millimetres daily, or at 2"26 metres per annum, and which 

 water, as Duclaux says, may remain six months or more as subterranean 

 water before appearing on the surface as springs, gives ample time for 

 sterilization. Exceptions to this sterilization of ground-water exist, as 

 may be seen in calcareous soils, in which the action of carbonic on the 

 chalk has caused by solution the formation of fissures often of great depth 

 and extent, thereby allowing the carrying of contaminations to great 

 distances. The conditions which regulate filtration in soils are, as Petten- 

 kofer remarks, sufficiently constant for general laws to be made regarding 

 them. He remarks that " the volume of the pores does not vary much in 

 different soils, and may be considered to occupy one third of the whole. 

 The dimensions of each pore may however vary considerably in different 

 soils. In soils which contain large pores the water percolates rapidly ; 

 a compact soil in which the pores are very fine is essentially hygrometric 

 or retentive. The level of a subterranean water may hence be uninfluenced 

 by the heaviest rainfalls, if the water stratum be deep and separated from 

 the surface by layers which may retain the heaviest rainfalls without 

 allowing a drop to pass through. It is apparent, therefore, that with 

 allowances for different kinds of soil, we have roughly different areas, or, 



