340 On the Leniting Point of Petroleum. 
same reason, doubtless, that gun-cotton may be ignited on 
a heap of gunpowder without setting fire to the latter; the 
heat evolved is intense enough, but the explosion occurs too 
quickly for the subjacent material to be ignited. Neither 
the petroleum nor the gunpowder, however, is any the less 
dangerous on that account; for, should the ignition of the 
mixture of petroleum-vapour and air cause an explosion so 
slight as only to force up a little of the petroleum into a fine 
splash of spray, ignition of the petroleum will inevitably 
ensue, as in Z#at condition petroleum is inflammable even 
at very cold temperatures. 
But to proceed to show why a given specimen of petro- 
leum may be observed to ignite at various temperatures. 
The zzme employed in the operation of taking the ig- 
niting point of petroleum as usually performed, is the first 
cause of variation to whichI shall allude. About two ounces 
of a specimen of petroleum marked “Ladoga;’ was placed in 
a small dish or capsule about five inches wide and one inch 
deep, and veryslowly warmed by a gas-lamp, a small flame 
being passed near the surface of the liquid every quarter or 
half minute. The liquid was stirred with the bulbed end 
of a thermometer, the temperature rising at the rate of 
about a degree per minute. In this way the experiment 
was carried on until at 124°, as marked on the stem of the 
thermometer, the petroleum-vapour ignited, and the liquid 
also ignited at the same instant. A similar experiment 
was performed on a fresh specimen of the same oil, except 
that the temperature was quickly raised; ignition of the 
vapour took place at 95°, and the oil itself took fire at 111°. 
It would seem that the highly volatile portion of the petro- 
leum, the portion which forms inflammable vapour, soon 
escapes when the oil is heated in an open vessel, leaving a 
liquid less inflammable than before. We know from the 
researches of Ronalds, Schorlemmer, Pelouze, Cahours and 
others, that petroleum, utterly unlike sperm, colza, olive, 
and the other old burning oils, is a mixture of a large num- 
ber of liquids of different degrees of volatility, and, indeed, 
containing bodies which when isolated are permanent 
gases, hence the above result might have been expected 
The rate of escape of vapour from the oil and from the 
operating vessel is another cause of variation in the igniting- 
point. Some of the same “Ladoga” oil was treated as 
already described, except that it was heated somewhat 
quickly, and was only stirred just before taking a reading 
of temperature. This time ignition occurred at 98°. 
