Miscellaneous, 



[JULY, 1911. 



be roughly parallel to the ground sur- 

 face. This is not to be wondered at, as 

 these strata have been formed in the 

 long process of time and at different 

 periods. 



Assuming that these conditions exist 

 in any particular place, and that some 

 of these strata are formed of porous 

 water-bearing material and others are 

 of a more or less watertight nature, and 

 again that some impervious stratum is 

 overlaid by a water-bearing soil, then 

 one would naturally select the hollows 

 in the ground surface rather than the 

 elevations as sites where well-sinking 

 would be likely to prove successful. 



This is another of those cases where 

 such a principle seems almost too axio- 

 matic to require statement, but the 

 writer has seen so many cases where 

 the exactly opposite course has been 

 taken, resulting in disappointment and 

 loss, that it may perhaps be worth 

 while to draw attention to the point. 

 The probability of finding water moder- 

 ately high up on the slopes of an ele- 

 vation from which the ground falls 

 away in every direction is so remote that 

 it is certainly not worth while spending 

 money on it. 



9. Let us now turn to the case of the 

 pure sand soil, the other extreme from 

 the bare rock. There is no better form 

 of storage reservoir— having regard to 

 conservation of the water supply— than 

 a deep bed of sand overlying a water- 

 tight stratum of clay or rock. Thus, in 

 Holland, the sand dunes along the coast 

 are a source of water supply for large 

 cities and towns, but there the sea 

 level takes the place of the underlying 

 rock. The water which reaches this 

 sandy bed, either directly in the form 

 of rain, or by flow from some higher 

 level, is retained there under the best 

 conditions, and subject to the minimum 

 of loss by evaporation and absorption, 

 properly so-called. 



The losses by percolation depend 

 largely on the relative area of the 

 sand bed, as compared with its depth 

 and the depth of water supplied in a 

 year, as replenishment, over the unit 

 area. Also on the slope, if any, of the 

 subsoil watertight stratum. 



Where large areas of sand deposit are 

 found they occur usually in level ground, 

 and the substratum is similar. An 

 exception occurs in the case of a deep 

 sand bed in a river. The watertight, or 

 partially watertight, substratum slopes 

 downwards similarly to the sand bed 

 surface, and when the water in the 

 river has ceased to flow over the sur- 

 face it still continues its travel down- 



wards under the surface. A good site 

 for a well can often be selected on the 

 margin of such a river, and the water 

 will be supplied by percolation laterally 

 from the bed. 



There is a certain amount of percol- 

 ation, and consequent loss thereby, in 

 even a very level sandy tract, as des- 

 cribed above, under any ordinary con- 

 ditions. If it be supposed that the 

 watertight substratum and sand surface 

 are absolutely level for a considerable 

 area, but that the substratum curves 

 upwards until it meets the surface of 

 the sand all round at some distance 

 from the centre of this area, then the 

 travel of the water which falls in rain 

 or is supplied otherwise over this area 

 uniformly will be ariested, and the 

 "water plane" or "surface of satur- 

 ation," say the level of the free surface 

 of water in wells sunk in various posi- 

 tions over the area, will be the same 

 in each place. Such a case is rarely 

 found, and there is usually some open- 

 ing on one or more sides possibly all 

 round by which the water can escape 

 by percolation. This process goes on 

 until the friction caused by the water 

 having to flow between the grains of 

 sand overcomes the action of gravity, 

 which tends to make the water flow, 

 and all motion is arrested. The water 

 plane will then be found to represent a 

 sloping surface, falling towards the out- 

 lets for escape, and varying in gradient 

 according to the character of the sand 

 as regards its coarseness or fineness. 



Supposing, therefore, that this sand 

 bed is 10 feet thick, and that the water 

 in a well situated near the centre of the 

 area in question stands at a level of 

 5 feet from the surface and 5 feet deep in 

 the well, this depth will be gradually 

 reduced to perhaps one foot or only a 

 few inches in wells sunk near the 

 margins of the area on the side or sides 

 on which the water can escape by 

 percolation. 



It stands to reason, therefore, that 

 the best site for a well in such positions 

 is one near the centre of the sandy tract, 

 or at least at the point or points most 

 remote from the outlets for escape. 

 (Diagram No. 3 illustrates this.) This 

 seems to provide a contradiction to the 

 remarks made previously about select- 

 ing a natural depression or hollow for 

 a well site, as in actual fact the centres 

 of these sandy tracts are frequently the 

 highest points, and the selection of a 

 well site at such places seems paradoxi- 

 cal. The subsoil conditions are, however, 

 the most important to keep in view. 



