164. Aprnry—Unrecognised Factors in the Transmission of Gases through Water. 
with boiling sea-water, and immediately closed air-tight with indiarubber corks. When the 
water had cooled, the tube was further exhausted by means of the mercury-pump. One limb 
was then uncorked, and the water in it was saturated with atmospheric gases by a current 
of air drawn through the surface-layer to a depth of 50 mm. for forty-eight hours. After 
this, a sample was drawn from the bottom of the open limb, and one from the lower portion 
of the closed limb, about 200 mm. from the bottom. The gases were extracted from these, 
and the nitrogen in each determined with the following results :— 
Open limb. Closed limb. 
12°12 200 600 560 LG ae 1:49 
Six months later another sample was collected from the lower portion of the 
closed limb, as before, and the nitrogen in it determined. It was 1°73. 
Experiments with broken surfaces. 
The experiment No. 1, and several others which were made during the earlier 
part of this investigation, but which it is unnecessary here to record, since the 
results were similar, shows that a retardation of solution takes place at the surface 
of the water exposed to the air. Experiments were accordingly made in which 
the surface of the water exposed to the air was kept continuously broken. 
No. 4.—A tube, 1100 mm. long and 50 mm. diameter, was filled with nitrogen-free sea- 
water, and unmersed in the water-jacket to a depth of 1025 mm. A glass stirrer of a 
gridiron-shape, 40 mm. broad, was fixed to dip 25 mm. below the surface of the water, and 
was mechanically rotated for nineteen hours. At the end of this time the temperature of 
the upper layer of the water-column above the water-jacket was 8°3°, and that of the water- 
jacket 7:8° C. Samples were drawn from the top and bottom layers of the water-column, 
and the nitrogen determined in each with the following results :— 
Top layer. Bottom layer. 
12°69 ae as. ee “ a 12°45 
At commencement, ... oe ce ce 1°51 
No. 5.—A similar experiment was made with distilled water with results as follows -— 
13°29, A SNE ere et 3 02 
At commencement, see aa ae 1:46 
Sea-water saturated at 8°3° C. contains 12:91 ce. N, per litre. Distilled water 
saturated at 8°3°C. contains 15:92 cc. N, per litre. 
In order to provide in these experiments the means of detecting whether or 
not the mechanical agitation of the surface-layers of the water-columns caused a 
circulation of the water to the bottom, a large crystal of potassium bichromate was 
placed at the bottom of the tubes; but in no case was the dissolved bichromate 
