Ox the Igniting Point of Petroleum, 341 
‘Stirring, as is well known, promotes the escape of vapour 
from a liquid, and in this case also by causing eddies and 
currents of air, promotes its escape from the surface of the 
liquid. 
"The form of the vessel in which the petroleum is heated, 
by accelerating or retarding the escape of vapour, also 
causes variation in the igniting-point. Some “ Ladoga” 
oil was heated gradually but quickly in a wide mouthed 
bottle; the mixture of petroleum-vapour and air thus 
formed in the upper part of the bottle gave indications: of 
ignition at 92°, and fairly burnt at 94°, several degrees be- 
low the igniting-point in an open vessel. 
The arrangement of the vessel over the source of heat of 
course influences the result. Ifa dish, &c., be so placed 
that its upper walls, not covered by the liquid, be- 
come much hotter than the lower portions, then the 
petroleum stirred up on to the hot part is quickly con- 
verted into vapour and ignites, while perhaps the liquid 
itself is comparatively cool. This is simply bad manipula- 
tion, and any one making an error of this kind should not 
be trusted with the performance of such experiments. 
Making the experiment in a deep glass beaker, or wide- 
mouthed glass bottle, and frequently introducing a large test 
flame, that the upper part of the vessel becomes full of the 
products of combustion to the exclusion of much ofthe air, 
would, of course, be equally bad manipulation. It is, per- 
haps, unnecessary to allude to these causes of variation, 
but that many of those to whom these remarks are addressed 
deal much in petroleum, but have no knowledge of practi- 
cal chemistry. The usefulness, therefore, of the observa- 
tions must be the excuse if they are trite and common- 
place. 
Shaking with aiy causes great variation in the igniting- 
point of a specimen of petroleum. A fresh specimen of 
the same refined “ Ladoga” oil was gradually warmed ina 
bottle, the specimen being violently shaken before the test- 
flame was introduced. In this way explosion actually took 
place at 78° F. This was an extreme experiment, but it 
serves to show the influence which more or less shaking or 
so to speak, washing out of the vapour from the oil by air, 
exerts on the point of ignition. It is true that when once 
in a lamp petroleum is not likely to be much shaken about, 
yet there are other circumstances under which violent agita- 
tion may take place ; it is only fair, therefore, that in taking 
the point at which a specimen of petroleum evolves in- 
