74 BULLETIN 1059, U. S. DEPARTMENT OF AGRICULTURE. 
soil wells, which has been thoroughly tried at the Fremont Experi- 
ment Station and elsewhere. 
At each station where soil moisture is to be determined periodi- 
cally, a well 18 to 24 inches in diameter and 4 or more feet deep may 
be dug. At the nearest available point a soil quarry is established 
for each station or group of stations having similar soils. From this 
quarry is taken soil containing only a moderate amount of humus, 
and which should be sifted through 4-mesh screen. At the outset 
sufficient sifted soil is obtained to fill the well and to furnish 1 or 2 
cubic feet of reserve, the whole being thoroughly mixed. The sifted 
soil should be firmly tamped into the well. It will be better if the 
well may be allowed to stand a year before being used, the soil be- 
coming settled by water action and being to some extent penetrated 
by roots. 
In order to maintain uniform conditions at the surface, each soil 
well should be kept free of litter by means of a frame of 1 by 4 inch 
boards, 18 or 24 inches square, which may be slightly sunk in the 
soil. Over this is placed a slightly larger frame covered with hard- 
ware cloth. It is evident that this frame will interfere with sur- 
face erosion. The surface of the soil in the well should at all times 
be kept flush with the surface of the ground around, so that the 
amount of water available for absorption is not appreciably greater 
or smaller than elsewhere. 
At the time of digging any soil well, samples of the native soil at 
1, 2, and 3 feet from the surface and other depths at which mois- 
ture is to be determined, as well as one of the prepared soil for the 
well, should be obtained for testing. Each sample should comprise 
about 30 pounds and should be air-dried, unless it is to be used imme- 
diately. 
Each soil quarry should be permanently designated at the time of 
its first use, and a record may be made of the quality and location of 
material taken therefrom, so that in the future fresh supplies for the 
well may be obtained, with a little trial, very nearly like the original. 
As such soil wells are used year after year, it will be noted that 
the finer material is to some extent concentrated in the lower layers, 
especially if the soil of the well is decidedly sandy and loose. This 
will not be found so important a change if the soil is compact, or 
contains considerable humus. 
The question naturally arises, how will the moisture in one of 
these wells compare with the moisture of the native and undisturbed 
soil on either side ? For the reason that the soil of the well is quite 
certain to be finer than the native forest soil, it is evident that the well 
soil will always contain a higher percentage of moisture. Further- 
more, without going into all the details, this question is answered 
unequivocally by saying that the moisture content, as determined 
