643 
ON THE TESTING OF PETROLEUM SPIRIT. 
BY F. C. CALVERT, F.R.S., F.C.S., ETC. 
Having read with much interest of late several papers on the Petroleum 
Act of 1868, on the uncertainty of getting good results when testing the 
u flashing-point ” of petroleum spirit (especially the papers published by Hr. 
Paul in the ‘ Chemical News,’ and a pamphlet by Mr. Norman Tate, Liverpool), 
I deemed it my duty to make a series of experiments, with the hope of throw¬ 
ing some light on the subject; the more so, that the defect of the Act of 1868 
is, that no length of time is specified for raising the petroleum spirit from 
natural temperatures to its flashing-point. Thus the Act states :—“ The outer 
vessel shall be filled with cold or nearly cold water. A small flame shall be ap¬ 
plied at the bottom of the outer vessel, and the thermometer shall be inserted 
into the oil, so that the bulb shall be immersed about one and a half inches be¬ 
neath the surface.” What is understood by a “ small flame,” appears to me a 
difficult question. Whether it shall be such as to require fifteen, twenty, or 
thirty minutes, to raise the petroleum spirit from natural temperatures to its 
flashing-point,—say 98° F. or 105° F., —the Act does not state. 
The following experiments, made with six samples of petroleum spirit placed 
in my hands by the magistrates of Manchester, will show how important the 
above question is with reference to the testing of petroleum spirit, and the ab¬ 
solute necessity, for the safety of the public as well as for the interests of those 
who deal in the article, that the Act should define the exact space of time that 
should be employed in the testing of any sample of petroleum spirit:— 
Sample of Petroleum 
Time, 
Time, 
Time, 
at 52° F. 
15 minutes. 
20 minutes. 
30 minutes. 
1 
96° F. 
98° F. 
102° F. 
2 
92° 
99° 
101 ° 
3 
90° 
98° 
101 ° 
4 
94° 
96° 
104° 
5 
96° 
98° 
110 ° 
6 
95° 
99° 
108° 
These results clearly show the influence of time in raising six samples of pe¬ 
troleum spirit from 52° F. to their flashing-points. Thus, when fifteen or 
twenty minutes are employed, the whole of the six samples could be called 
petroleums, according to the Act of 1868, consequently the owner would be 
liable to a penalty and the loss of the fluids ; whilst, if the time employed to 
heat the liquid is half an hour, they would all be considered not petroleums, their 
flashing-points being above 100° F. 
Therefore, there can be no doubt that according to the time employed so are 
the results altered ; the longer an operator is in completing his test, the higher 
will be the flashing-point of the spirit. This is probably due to the fact, that 
the most volatile products gradually escape into the atmosphere, and are never 
in sufficient quantity to produce a flash when a taper, as described in the Act, 
is passed within a quarter of an inch from the surface of the spirit. I am, 
therefore, of opinion, that as the Act has been made to protect the public 
against fire and explosions resulting from the employment of too highly in¬ 
flammable hydrocarbons, the chemist or person called upon to test liquids of 
this class, should raise the temperature of the liquids as quickly as possible (of 
course employing the apparatus described in the Act), otherwise they favour 
the vendor and manufacturer to the detriment of the consumer. 
The next series of experiments were made with a view of corroborating a 
statement made by Mr. Norman Tate, viz. if two thermometers are placed into 
the petroleum spirit ; one, as indicated in the Act, one and a half inches below 
