310 Unversity Geological Survey of Kansas. 
THE METHOD WHICH WAS EMPLOYED IN THE COMPARATIVE STUDY OF 
THE PROPERTIES OF KANSAS OILS. 
About 0.7 liter of crude oil was weighed in a tared liter dis- 
tilling flask and subjected to fractional distillation, the collecting 
vessels being changed at the following temperatures C.: 125°, 
150°, 200°, 250° and 300°. In each case after first reaching the 
temperature at which a fraction was to be made the Bunsen 
flame was removed until the thermometer indicated a fall of 25°, 
when the distillation was renewed until the dividing tempera- 
ture was reached a second time, after the method of Engler. 
Hach fraction was weighed, and it was found that the loss of 
weight of oil in the flask exceeded the sum of the weights of the 
distillates. This difference in weight was due to the volatiliza- 
tion of the dissolved gases contained in the crude oil, and 
amounted to 0.3 to 1.5 per cent. 
Distillation was then carried on in a vacuum, the pressure 
generally being reduced to that corresponding to 20 mm. of 
mercury. Fractions were collected at intervals of 25° C. The 
weight of the vacuum distillates was found to equal the loss of 
weight in the flask; showing that no appreciable amount of gas 
was formed through cracking. 
The residue from the vacuum distillation was transferred to 
a half-liter distilling flask, which was supported on an iron 
stand and protected from drafts by two asbestos dishes placed 
vertically, one in front, the other behind, about half an inch 
from the walls of the flask. Heat was supplied by a triple 
burner, moderately at first, but finally the entire flask was sur- 
rounded by flame. The delivery tube of the flask was con- 
nected to an air-condenser. The receiver was provided with a 
side neck, through which the gases formed during cracking 
were passed either to a gasometer or to a Bunsen burner, 
where they were burned to get rid of their disagreeable odor. 
In general, the weight of gases and volatile products lost dur- 
ing the cracking process was approximately equal to the weight 
of the coked residue. 
The samples of oil were collected in accordance with the 
directions of Professor Haworth, the state geologist, and every 
effort was made to have them truly represent the oil as it is 
obtained fresh from the wells. However, they were collected 
during the summer months, and it is probable that some of 
them, while exposed to the heat of the sun upon the platforms 
at the railroad stations and express offices, may have lost con- 
