1897.] COMPOSITION OF AMERICAN PETEOLEUM. 129 



variably much decomposition. To illustrate the effect of air at high 

 temperatures on distillates with high boiling points, in a course of 

 distillations in vacuo of Russian crude oil, accidentally air was 

 allowed to enter a still in which distillation was proceeding without 

 decomposition under a tension of 50 mm. at 250''. As soon as the 

 air came into contact with the hot vapors, there was a violent ex- 

 plosion sufficient to send the thermometer out of the still and shatter 

 it against a brick wall several feet distant. Air let into a still under 

 similar conditions in which Pennsylvania oil is distilling usually 

 causes flashes of light, but no explosion. From these observations 

 it is evident that the advantage of distillation in vacuum depends as 

 much at least on exclusion of air as on the reduction in temperature. 



Definite statements relating to the composition of petroleum from 

 diff'erent American fields must, at present, be limited to the distil- 

 lates below 216°. But so far as it is possible to draw conclusions 

 from data collected there seems to be no possibility of distinction 

 based on geological occurrence and composition. High percen- 

 tages of sulphur constituents are usually associated with limestone 

 formations as the source of occurrence of the crude oils. But study 

 of the varieties of crude oil from widely different sources leaves no 

 basis for this distinction. A petroleum from South America occur- 

 ring in a system of shales and sandstones contained 0.70 per cent, 

 of sulphur (Mabery and Kittelberger, Proc. Amer. Acad., xxxii, 

 185). Another variety from Oregon having no connection with a 

 limestone formation also gave a high percentage, 1.19 per cent of 

 sulphur. A specimen of petroleum from Japan, now under exam- 

 ination in this laboratory, gave 0.5 per cent, of sulphur. The im- 

 mense deposits of petroleum in Roumania occurring in shales and 

 sandstones are mostly high sulphur oils. No distinction can there- 

 fore be based on sulphur contents and geological occurrence. 



It seems doubtful whether a distinction can be based on specific 

 gravity and geological occurrence. The Pennsylvania oils differ 

 from most ethers in their low specific gravity, varying in the main 

 between 0.80 and 0.82 ; such light oils as the amber variety from 

 Berea grit sandstone is as low in specific gravity as o. 79 (Mabery 

 and Dunn, American Chemical Jour?ial, 1896, 11). The limestone 

 oils are higher, those from the Trenton limestone giving 0.82 to 

 0.85, and those from the Canadian Corniferous limestone 0.85 to 

 0.88. But the Russian oil, the South American oil mentioned 

 above, the Japanese oil, and the Roumanian oil all show a high 



