230 University Geological Survey of Kansas. 
general account of the work done on this gas is taken from a 
paper by Cady and McFarland :1!° 
“A sample was sent to the University of Kansas for analysis, 
and was found to contain only a small proportion of combus- 
tible constituents and a very large proportion of incombustible 
residue, such as is usually reported in gas analyses as nitrogen. 
The results of the analysis were published in a paper by Ha- 
worth and McFarland,!! and by McFarland.!2 When the ni- 
trogenous residue, mixed with oxygen, was subjected to the 
action of an electric spark over a strong solution of potassium 
hydroxid for a very long time, there was left a residue which 
refused to combine with the oxygen, and was evidently not 
nitrogen. This suggested the presence of some of the inert 
gases of the argon group. 
“Further investigation of the Dexter gas by the methods de- 
scribed in this paper has shown the presence of considerable 
- quantities of helium, and in a paper read for us by Prof. E. H. 
S. Bailey, at the New Orleans meeting of the American Chem- 
ical Society, we reported 1.84 per cent. helium. Since that 
time the investigation has been extended to many other Kan- 
sas natural gases, and also to gases from other fields, and in 
nearly every case helium has been found. 
‘‘Along with the determination of helium, an extended an- 
alysis of each gas has been made to determine its other more 
common constituents: 
These analyses have revealed an interesting variation in 
composition with locality, which is discussed more in detail in 
the latter part of this report. In attempting to compare these 
gases with those from other fields, we have felt the lack of any 
considerable collection of natural gas analyses in the literature. 
The difficulty which we have experienced in finding this data, 
scattered as it is through a great mass of scientific journals 
and official reports, has led us to add to this paper a somewhat 
extended series of abstracts of analyses gathered from all 
available sources, together with references to the original pa- 
pers. It is not claimed that these cover all of the work done on 
natural gas analysis, even in this country. Many analyses of 
gases from volcanoes, mineral springs and mines have been 
omitted intentionally, as not being quite in the same category 
as the natural gas used for industrial purposes. Enough of 
these have been given, however, for comparison, and it is 
110. Jour. Am. Chem. Soc., 29, 1523. 1907. 
Will, SelenG, TOL Wily oO. Wik 
112. Trans. Kan. Acad. Sci., vol. 19, p. 60. 
