CHEMICAL AND PHYSICAL PAPERS. 57 



SOME KANSAS PETROLEUM. 



By Edward Bartow and Elmer V. McCollum, University of Kansas. 

 Read before the Academy, at Topeka, January 1, 1903. 



OUR attention was called to the fact that there must be consider- 

 able variation in Kansas petroleum when some specimens dif- 

 fering entirely in character were sent to the laboratory of organic 

 chemistry for analysis. We therefore thought it advisable to collect 

 specimens from various parts of the Kansas oil fields, and to make 

 tests to prove our conclusion that there was a decided variation in the 

 character of the oils from different parts of the field. This paper tells 

 of our observations of the specific gravity, flash point and burning 

 point of eight samples of oil collected in Allen and Neosho counties. 



To prevent loss of volatile oils, it is desirable that the samples be 

 collected directly from the wells. Owing to the different stages in 

 the development of the pumping arrangements of these wells, it was 

 not always possible to collect the samples under like conditions. We 

 have noted the conditions of collection below. 



We have not included the record of the depth of the wells in this 

 paper, because the height above sea-level and the dip of the strata 

 must also be taken into consideration in order to make an accurate 

 comparison of the depths. 



Sample No. 1 was from a well about three miles north of Chanute. 



Sample No. 2 was from a well about one and one-half miles east of 

 Chanute. 



Sample No. 3 was from a well three miles south of Humboldt. 

 This sample was taken directly from the well by the writer. 



Sample No. 4 was from a well two miles north of La Harpe, and 

 had been in barrels exposed to the air for several days before collec- 

 tion was made. The well was new, and had been shot only about ten 

 days. 



Sample No. 5 was from a well in the northern part of the city of 

 Humboldt, and was taken from a tank into which it had been pumped. 

 The Engler test showed water. 



Sample No. 6 was from a well four miles southwest of Humboldt, 

 and was in part taken directly from the flowing well and part from a 

 barrel. 



Sample No. 7 was from a well about a mile west of Humboldt, and 

 was pumped directly from the well. 



Sample No. 8 was from the schoolhouse well in Humboldt. The 

 well is primarily a gas well, but some oil collects and must be pumped 



