CADY AND MCFARLAND.| Composition of Natural Gas. 265 
cooled charcoal for some time, the helium and neon are removed 
by strong pumping with the mercury-pump, which is kept going 
until helium can no longer be detected by the spectroscope and 
the vacuum is very high. 
Then the charcoal bulb is allowed to warm up, thus allowing 
the argon to pass out of it. The lines of argon immediately ap- 
pear in the spectrum and are identified by careful measurement 
of their wave-lengths and by comparison with those from 
known specimens of argon. 
The tedious nature of this operation, and also that for the de- 
tection of neon, have prevented us from testing many gases for 
them. It may be remarked, however, that neon is a usual con- 
stituent of the gas furnished to the University through the 
pipe-lines of the Kansas Natural Gas Company and has been 
several times extracted from this source. Argon is present in 
the Dexter gas. It is probable that both gases will be found as 
normal constituents of most of the samples of natural gas, if 
they are tested for. 
KRYPTON AND XENON. 
We have in hand the concentration of a considerable amount 
of Dexter gas, and propose, when a sufficient amount of the 
inert residue has been obtained, to test for krypton and xenon. 
It is our expectation that by means of cocoanut charcoal bulbs 
held at the proper temperatures the separation of the entire 
series of gases of the argon group may be effected. 
HYDROGEN SULFID. 
This gas was tested for in several of the gases examined by 
passing the gas for several hours through solutions of lead 
acetate and copper sulfate. In no case was any precipitate pro- 
duced, and from this fact and the fact that no odor of hydrogen 
sulfid has been detected in any of the gases examined, we con- 
clude that hydrogen sulfid is not present, except in the very 
faintest traces. 
ANALYTICAL RESULTS. 
The results of the analyses are given in table 27. The an- 
alyses are arranged in the order of the quantity of helium con- 
tained in the gases. A statement of such data as were ob- 
tainable regarding the location, depth and pressure of the wells 
producing the gases, as well as other matter of interest regard- 
ing their occurrence, is appended to the table. 
