SUBTERRANEAN GASES. 
255 
with water and its repeated filling and emptying by siphon insured a corresponding 
quantity of dry and pure gas in the dry bottle. 
The two samples from the Conundrum mine, collected under the conditions 
explained on page 262, were analyzed by the department of chemistry of Cornell 
University through the kindness of Prof. L. M. Dennis, the analyst being Mr. L. F. 
Hawley. 
Volumetric analysis of gas from Conundrum mine. 
• 
1 . 
2. 
Carbon dioxide. 
10.2 
8.3 
Oxygen... 
5.7 
10.2 
Nitrogen (by difference). 
84.1 
81.5 
100.0 
100.0 
No methane or other hydrocarbons were found, nor was there any carbon 
monoxide or hydrogen. 
One sample from the Elkton mine, level 7, south, collected by Mr. L. C. Graton 
on April 15, 1904, under conditions explained above, was also forwarded to Professor 
Dennis, who reports on it as follows: 
The gas was received in a 2-liter glass-stoppered bottle, stopper tied in and protected with paraffin-coated 
canvas. 
The analysis was made by Dr. Arthur W. Browne, of the department of chemistry, Cornell University. 
All determinations of absorbable constituents were made over mercury in a Hempel burette provided with a 
water jacket . The absorbents were contained in mercury pipettes that were charged with not more than from 
15 to 25 cc. of the absorbent. In the absorption analysis the first series of results was rejected to avoid errors 
due to incomplete saturation of the reagents with the nonabsorbable constituents. The percentage of water 
vapor was determined in two fresh samples of the gas by absorption with concentrated sulphuric acid, the acid 
being of course saturated with the other constituents of the gas. Nitrogen was determined by sparking the 
unabsorbed residue with an excess of oxygen over a solution of potassium hydroxide. This was first carried 
on in a Winkler-Dennis combustion pipette fitted with an adjustable spark-gap in place of the spiral coil. 
The final experiments, looking to the removal of nitrogen, were made in a Hempel explosion pipette with the 
spark gap slightly widened. When the residue was sparked in this manner with oxygen, contraction took 
place and continued for about forty hours. The mixture was sparkled for several hours after contraction had 
ceased. The spectrum was then observed and nitrogen lines were found to be absent. The excess of oxygen 
was then removed and the inert residue was measured. Spectroscopic examination of this residue revealed 
the presence of argon only. This result was confirmed by comparison (by superposition) with the spectrum of 
an argon tube. The results of the analysis are as follows: 
Volumetric analysis of gas from the Elkton mine. 
Water vapor. 1-4 
Hydrocarbon vapors. 0. 0 
Carbon dioxide. 14. 7 
Heavy hydrocarbons. 0. 0 
Oxygen. 5. 6 
Carbon monoxide. 0. 0 
Hydrogen. 0. 0 
Methane, etc.. 0. 0 
Nitrogen. 76. 8 
Argon. 1.5 
100.0 
I 
This gas may be considered as a mixture of about 25 per cent air, 59 per cent 
nitrogen and argon, 15 per cent carbon dioxide, and 1 per cent water vapor. There 
is considerably more argon present than in normal air. If the air be regarded as an 
accidental admixture the pure gas would consist of about 20 per cent carbon dioxide 
and 80 per cent nitrogen and argon. 
