66 The Colorado Experiment Station. 
amounted to 0.333 percent of the air-dried sample. No. 999 
is the residue obtained by evaporating a sample of the water taken 
from the cellar, August 1910, to dryness. The residue equalled 
996.1 grains per imperial gallon or 14,260 p. p. m., while No. 
1040 is a similar sample taken December 28, 1910 carrying 1229.27 
grains per imperial gallon or 17,561 p. p. m. These samples can¬ 
not be taken as actually representing the ground water owing to 
the fact that concentration due to evaporation must have taken 
place. The dimensions of this cellar as stated are 25x50 feet and 
while the house is built over it the cellar windows were open and 
we are uncertain in regard to the changes in it principally due to 
concentration. The first water observed in this cellar came in in 
the winter of 1908-1909 or about eighteen months before the 
sample 999 was taken and the analyses present the following re¬ 
sults : the white alkali which had effloresced contained no nitrates; 
the white alkali mixed with some soil from slightly higher ground 
contained some nitrates, one twenty-fifth of the weight of the water- 
soluble ; but the soils which would ordinarily be considered as free 
from alkali contain nitrates equal to one-quarter of the weight of 
the water-soluble. 
The present owner of this land has put in upwards of 15,000 
feet of tile drains. I saw about 7,000 feet of the trenches open in 
the very worst part of the land. The surface of the land was 
muddy as though they had recently had a heavy rain or the land 
had just been irrigated, the second spit was decidedly dryer than 
the first and while there was some water in the bottom of the trench, 
here and there filling the depressions made by uneven digging, by 
far the largest portion of it was entirely free from running water. 
I was very much surprised at the small amount of water appearing 
in so great an aggregate of trenches. It should, however, be 
stated that water flows out of such land with exceeding slowness. 
The fact was that the very surface of this bad land was the wettest 
portion of it and my impression is that this was always the rule 
with it. The land at the north end of the orchard is dry and, as 
previously stated, I was informed by two men, both interested in 
the success of the work, that the drain trench, from four to six 
feet deep, developed no water whatever. The effect of ground 
water upon the trees depends upon the depth at which the roots 
actually feed as well as upon the heighth of the water plane and 
the capillary power of the soil. I saw this trench when it was 
partially dug and I called the owner’s attention to the fact that 
it was rarely the case that the roots were more than ten inches 
deep, though we occasionally found them, small fibrous roots, four¬ 
teen inches deep. The manager informs me that some of the trees 
in this section of the orchard burned in the autumn of 1910 which 
seems to he very probable as they are practically dead in the 
