84 TRANSACTIONS OF THE [jAN. 31, 



me that, of the fifteen hundred or two thousand wells selected 

 and driven by himself, not more than two or three have ceased to 

 act, and that in nearly all the supply remains undiminished. 



Experiments were made on Long Island to determine the line 

 of depression of the water-level in the ground immediately adja- 

 cent to the plant of one hundred wells, while the pumps were 

 drawing to their full capacity. 



Test wells were sunk just outside of the plant on four sides, 

 and a line of test wells extended out to a distance of 4,000 feet 

 from the pump-house. Observations were made during three 

 months. At the distance of 4,300 feet, the water-level was re- 

 duced about 6 inches. At a distance of 2,300 feet, about 2 feet 2 

 inches, and in the immediate proximity of the wells, within a 

 distance of 300 feet, the depression was 4 feet 8 inches. When 

 the pumps were stopped, all the levels were restored. 



The rate of flow to wells sunk in sand or gravel depends, of 

 course, upon the compactness of the material of the water-bear- 

 ing stratum, the resistance to the flow being proportional to the 

 narrowness of the minute channels through which the water must 

 flow, and to the number of windings and turnings which a 

 particle is forced to take throughout its course to the pumps. 

 The velocity of a particle of water running into an ordinary fil- 

 tering gallery along the margin of a stream is generally less than 

 one foot an hour. The velocity of a particle flowing towards the 

 pumps in the driven-well plant on Long Island, taken a little 

 outside the rectangle formed by the hundred wells, is about the 

 same — a velocity too small to be directly perceptible to the eye. 

 This small velocity shows what great relative resistances are 

 encountered. 



The form of the water-surface will, therefore, depend almost 

 entirely upon the coarseness or compactness of the materials 

 through which the water passes, and is not to be taken as a 

 measure of the degree of exhaustion of the source of supply. 



In all the explorations thus far made, it has been found that 

 the gravel-beds, available for water, are not continuous over 

 indefinite areas, nor of uniform thickness, but certain beds and 

 particular channels in certain beds appear to constitute the 



