1903.] MABERY—THE COMPOSITION OF PETROLEUM. 39 
Warren had shown the existence in Pennsylvania petroleum of the 
two series of isomeric hydrocarbons from pentane to octane. The 
large deposits of petroleum in northern Ohio and Indiana, Texas, 
Colorado, Wyoming and Kansas had not then been discovered. 
In 1885, soon after the first well was drilled that yielded oil from 
the Trenton limestone, two oil inspectors brought me a five-gallon 
can of Trenton limestone oil and remained while I examined it for 
them. This was my first acquaintance with the sulphur petroleums. 
I recognized at once the large percentage of sulphur and soon after- 
»ward began a study of the sulphur compounds. I am not now fully 
satisfied as to the nature of those sulphur compounds. Not long after- 
ward I also procured the Canadian sulphur oil, and carried along 
together the study of these products, the one from Trenton lime- 
stone and the other from the Canadian Corniferous limestone. The 
composition of Ohio oil has only recently been determined, with 
respect to the principal series of hydrocarbons, by a research com- 
pleted during the present month, seven years after it was begun. 
At first I had no preconceived ideas as to the series of hydro- 
carbons which compose these crude oils, except what knowledge I 
had gathered from the work of my predecessors ; but after the work 
had progressed far enough to see that the crude oils from the differ- 
ent fields were essentially different in certain constituents, especially 
in sulphur, I was inclined to look on the Trenton and Corniferous 
limestone crude oils as a special species, the sulphur petroleums, 
and to agree with Peckham in his specific classification of the dif- 
ferent petroleums as varieties of bitumens. But I soon became 
convinced that no such sharp distinctions based on composition 
could be drawn. Now after these years of arduous labor I have 
reached the conclusion that petroleum from whatever source is one 
and the same substance, capable of a simple definition—a mixture 
in variable proportions of a few series of hydrocarbons, the product 
of any particular field differing from that of any other only in the 
proportion of these series and the members of the series. I arrived 
at this conclusion only one year ago, when it was found that the 
higher distillates from Pennsylvania petroleum contain the 
series C,H,,_., which until then I had supposed was only to be 
found in the heavier California arid Texas oils, or the so-called 
asphaltic oils. Results obtained within the last two months show 
that Ohio petroleum has a similar composition. 
In support of this definition I would suggest that, so faras known, 
