New York Agricultural Experiment Station. 103 
and the bottle filled with spring water, after which it was replaced, and 
the air bubbles beneath it were let out by inserting a pin between the cork 
and the neck of the bottle, so that the vessel was completely filled with 
the water. The bottle was then placed under the receiver, and the air 
pressure reduced to seven pounds per square inch. The water imme- 
diately rose in the tube to the height of rather more than five-eighths 
of an inch, indicating an expansion of the air contained within the water. 
On admitting the air again, the liquid receded to its former height. 
These results seem to prove that changes in atmospheric pressure 
affect percolation. During the past season an investigation has been 
made of the flow of water from farm drains, with especial reference to 
this question. A general, though not an exact coincidence, was readily 
observed between the fluctuations of the barometer and of the drainage.* 
EXPEKIMENTS ON CAPILLARY ACTION IN SOILS. 
1. The height to which water rises by capillarity in different soils, 
and its rate of progress. 
2. The influence of matters in solution upon the height to which 
water rises by capillarity. 
3. The rate of capillary flow through different soils. 
4. The influence of matters in solution upon the rate of capillary 
flow. 
The Height to Which Water Rises by Capillaeity in Different Soils, 
and its Rate of Progress. 
Four glass tubes, each five feet long and very nearly one and five- 
eighths inches inside diameter, were supported in a vertical position 
by means of a stand constructed for the purpose. The lower end of 
the tubes entered a copper pan, in which distilled water was kept at 
a very nearly constant depth by means of an inverted supply bottle. 
The samples of soil were retained in the tubes by covering the lower 
end of the latter with fine brass wire-cloth, over which was placed a 
circular piece of filter-paper of a diameter corresponding to that of 
the tube, the wire cloth being kept in position by bending it upwards 
about the tube, and winding it tightly with fine copper wire. Thus 
the filter-paper prevented the passage of the soil through the wire 
cloth, while it permitted free access of water. 
Tube No. 1 was filled with muck, No. 2 with garden soil, No. 3 with 
sand and No. 4 with clay, all the soils being air-dry and sifted through 
* An illustrated article contributed by the writer to Agricultural Science, 
Vol. I, pp. 173-177, gives a full report of this investigation. 
