July 21, 1923 
Oxygen-Supplying Power of the Soil 
135 
conditions change (otherwise the rate of absorption of oxygen could not 
be taken as a measure of the ability of the soil to supply this element); 
(2) the oxygen absorbed must be removed and collected in some way 
so that the amount taken in during a test period may be quantitatively 
determined. 
After some preliminary experimentation a porous-porcelain cylinder 
closed at one end, such as is used in atmometry, 6 was adopted as a prom¬ 
ising form of absorber. To prevent the streaming of the gas through 
the porous wall the pores were finally filled with paraffin oil (Nujol). 
As experience has shown, oil-impregnated porous porcelain seems to be 
adapted for the taking up of oxygen by diffusion. The cylinder was 
ground down until the lateral wall, excepting the enlarged rim, was from 
1 to 2 mm. thick; the rim was then coated with sealing wax to prevent 
oxygen absorption in that region. The oil treatment was applied by 
filling the hot cylinder (about ioo° C.) and allowing it to stand upright 
for 12 hours, after which the superfluous oil was removed. 
The open end of the cylinder was closed by a rubber stopper holding 
two glass tubes, one of which extended through the cylinder nearly to 
the closed end, while the other terminated just inside the stopper. All 
joints were sealed with spar varnish, as was also done throughout the 
remainder of the apparatus which is described in the following paragraphs. 
Oxygen from the surroundings diffuses through the wall of the absorbing 
cylinder, tending to make the partial pressure of oxygen in the cylinder 
cavity equal to the external partial pressure in the immediate neighbor¬ 
hood. De-oxygenated gas (illuminating gas thoroughly scrubbed by 
alkaline pyrogallol solution, adopted merely because it was convenient 
and served well for these preliminary experiments, which did not deal 
with living plants) is passed slowly—60 cc. per hour—through the 
absorber, so that oxygen is removed from the cylinder as rapidly as it 
diffuses in from the outside. The internal partial pressure of oxygen is 
therefore always almost nil, while the instrument is in operation. The 
gas coming from the absorbing cylinder is conducted to a bottle contain¬ 
ing the indicator solution, in which the oxygen is all absorbed and 
measured. 
The indicator bottle is of the ordinary wide-mouth form, 4 cm. in 
diameter and 10 cm. high, closed by a sealed-in rubber stopper, through 
which pass the glass tubes for the entrance and exit of the gas. The 
entrance tube reaches to the bottom of the bottle, while the other extends 
only slightly beyond the stopper on the inside. 
Ten cc. of the indicator solution is placed in the indicator bottle. 
For this solution, which collects the oxygen coming in the gas stream 
from the absorbing cylinder, and which gives indication of the amount 
of oxygen thus collected, an alkaline solution of pyrogallol is used. It 
is prepared as follows: Two stock solutions, one containing 50 gm. of 
pyrogallol dissolved in 100 cc. of distilled water, and the other having 
450 gm. of potassium hydroxid in 500 cc. of distilled water, are mixed 
with oxygen-free water in the proportions of 1:2120. 
As is well known, an alkaline solution of pyrogallol absorbs elementary 
oxygen with great avidity, the absorption proceeding with marked 
rapidity and continuing when* the gas is rapidly supplied until large 
amounts have been absorbed. If made under proper conditions, such a 
pyrogallol solution is practically colorless before any considerable amount 
• Livingston, Burton Edward, the relation of desert plants to soil moisture and to evapora¬ 
tion. 78 p., 16 fig. Washington, D. C, 1906. Literature dted, p. 77-78. (Carnegie Inst. Wash. Pub. 50.) 
