40 ENGINEERING FOR LAND DRAINAGE. 



drain and well B. In the well two feet from the drain the 

 water stood . 3 foot above the top of the tile and in the 

 other .45 foot above. The profile would present, there- 

 fore, a more or less curved contour, convex upward." 



Prof. King, under whose direction the observations 

 were made, draws the following conclusions : 



' ' Assuming the water level at the several lines of 

 tile to be flush with the tops of the tile, and regarding 

 the water surface as presenting a right line of section, 

 the mean gradient for the ground-water surface plane 

 of saturation would be i foot in 25.3 feet. There were 

 other wells sunk outside the range of tile drains for the 

 purpose of ascertaining the height of the ground- water 

 above the water surface of the lake. Of these he says : 

 *'In well 29, 150 feet from the lake shore, the water 

 stood 7.2 feet above the level of the lake on June 27, 

 1892, and this would give a gradient of i ft. in 20.8. 

 In the case of the well at the Hall to which I have re- 

 ferred as having a water level 52 feet above the lake, 

 and situated about 1250 feet from the shore, the mean 

 gradient would be i in 24.04. In the fall of 1888, 

 Sept. 10, when the water level in the wells could not 

 have been affected by lateral percolation, the gradient 

 between well 29 and the lake was i in 35.8." 



He observes further that the water line of the tile- 

 drained field under consideration did not remain as de- 

 scribed, but as water was carried away by the drains, 

 the line was drawn down at a uniform rate, falling fast- 

 est on the highest ground where the level was highest. 



These researches serve to indicate in a tangible way 

 the relation of the line of saturation to underdrainage. 

 The method of averages used by Prof. King in drawing 



