628 
Journal of Agricultural Research voi. xxvii. No. * 
other so closely as to be almost indistinguishable. In the present experi¬ 
ment these soils were pulverized and passed through a sieve having holes 
2 mm. in diameter. The soil was then placed in glass tubes half an inch 
in diameter and 5 feet in length. The lower end of the tube was fitted 
with a perforated rubber stopper which was covered with a layer of 
absorbent cotton. As the soil was poured in, the tube was jarred in such 
a way as to make the soil settle firmly. The tubes were filled to the 4-foot 
mark and clamped to a stand in a vertical position. Distilled water was 
then poured into the top of the tubes to the depth of about 6 inches, and 
more was added from time to time, as needed. The perforated stoppers 
in the tubes permitted the escape of air as it was displaced by the descend¬ 
ing water. The line of demarcation between the wet and dry soil was 
very sharp. 
Table VI shows the rate of penetration of the water into the soil in the 
two tubes. In the tube containing soil No. K82 the water penetrated to 
the depth of 33 inches in three hours, while with soil No. 337 the water 
had gone down less than 4 inches at the end of the first three hours. In 
this tube, No. 337, a constant supply of water was kept on the soil for 
more than a year and the depth of penetration was recorded from time 
to time. At the expiration of 14 months the water had soaked down to 
the bottom of the soil column, or 48 inches. 
Table VI .—Penetration of water in two samples of dry pulverized soil in glass 
tubes 
Depth of penetration (inches). 
Time elapsed. j 
Soil No. K82. 
Soil No. 3, J 
1 hour. 
16. 7 
25. 0 
32. 7 
2. 5 •. 
3 - ^ 1. 
3 - 7 1 .- 
4 - 1 :. 
4-4 i. 
4.5 j -•••••-■ 
5.6 | 5-6 
9-3 i - 4 i 
12. 6 | . 165 
16. 6 | . 133 
20. 0 j . 113 
23. 1 1 . 103 
25. 5 { . 080- 
44. 0 1 . 086 
48. 0 ] . 066 
2 hours. 
3 hours.. 
4 hours. 
5 hours.... . 
6 hours. 
1 day. 
10 days. 
30 davs... 
60 days. 
90 days. 
120 days. 
150 days. 
12 months... 
14 months. 
Toward the end of this experiment it became difficult to determine 
the limit of penetration accurately. Instead of a sharp line of demar¬ 
cation between the wet soil and the dry, there was a gradual transition 
of color. Evidently the penetration of water was retarded by conditions 
in the mass of moist soil in the upper part of the tube rather than by 
any definitely impermeable area below. It was observed also that the 
rate of penetration of the water declined rather uniformly as time went 
on. This has been noted also in other similar experiments. 
The outstanding feature of this experiment is that with these two 
soils, closely similar in physical texture, there is a very great difference 
