346 
Journal of Agricultural Research 
Vol. XI, No. 7 
ing about the middle of the month. The rise recorded at the end of 
June was apparently checked in the wells nearest the pump, but did 
not reach the maximum in those farther away until the end of the first 
week in July. An irrigation in July caused a sudden rise followed by 
a decline which by August io had reached a point as low as any during 
the season. The stopping of the pump from August i to August 4 did 
not check the receding water table. During June, 1915, when the 
pump was not in operation, the water table averaged lower than during 
Fig. 6 .—Ground-water curves for 1914 and 1915 (wells 13,14. and 15). 
June, 1914, while in July after the pumping was resumed, the water 
table averaged higher than during the same month of the preceding 
year. Well 4, which was very near the pump, was affected more than 
any of the others by variations in the pumping. After August 10 the 
curves for the two years coincide quite closely. 
Figure 9 shows the water table for March, 1915, on four lines through 
the sump. The pump was not yet started, and the water table was 
