348 PROCEEDINGS OF THE AMERICAN ACADEMY. 



filter, and washed with a mixture of fusel oil and alcohol. The pre- 

 cipitate ou the filter was then placed in an extraction apparatus, and the 

 solid extracted with benzol. After evaporation of the benzol, the residue 

 was heated to 140' for one hour to remove the last traces of fusel oil. 

 19.8 grams of a greenish black solid remained, melting at 32°, equivalent 

 to 39.6 per cent of tiie heavy residue or to 1-4 per cent of the original oil 

 before evaporation. 



The solid was again dissolved in fusel oil and precipitated by alcohol, 

 giving a smaller quantity of solid melting at 45°. This treatment re- 

 peated gave a light brown solid melting at 57°. The latter was then 

 dissolved in ether and precipitated by alcohol, which gave a lighter 

 colored solid ; another repetition gave a white solid melting at 61°. Its 

 specific gravity was 0.7966 at 70°, corresponding to that of paraffine, and 

 a combustion gave percentages of carbon and hydrogen required for 

 paraffine hydrocarbons. 



0.1511 gram of the solid gave 0.4730 gram CO., and 0.1984 gram IL.O. 

 C 85.37 



H 14.69 



The specific gravity, melting point, and composition of this substance 

 indicate that it was composed of paraliine hydrocarbons, and demon- 

 strate the presence of the solid parafiine hydrocarbons in Pennsylvania 

 petroleum. 



Parafiine collected from heavy distillates in the refinery is therefore 

 obtained as it exists from the original crude oil, and it is not formed 

 during distillation. Indeed, as explained in another paper, it cannot be 

 formed ; for any decomposition of petroleum hydrocarbons by heat re- 

 moves hydrogen with a conversion of the hydrocarbon into a lower 

 series; for example, of the paraffine hydrocarbons of the series Cnllon+o 

 into the series C„Ho„, C„H2n_25 etc., and perhaps also forming the metliy- 

 lenes and condensed methylenes. Such changes do not involve an ab- 

 sorption of oxygen into the hydrocarbon molecule, but a removal of 

 hydrogen by oxygen as water. That oxygen compounds of the nature of 

 phenols are found in petroleum from various fields has been fully demon- 

 strated independently by Richardson and by Mabery (data not published), 

 and it is quite probable that most petroLuiu contain these bodies to a 

 greater or less extent ; but they have been formed by natural processes. 



In ordinary refinery distillation, there is probably little decomposition 

 below 300° C, but at this point cracking begins and increases rapidly to 

 the end. There are doubtless reactions at higher temperatures between 



