TRANSACTIONS OF SECTION B. 777 
few years later, and consists in passing the mixture through a water-valve into a 
tube heated to a known temperature. But Mallard and Le Chitelier’s second 
method is a still further improvement. In this they introduce the mixture into a 
heated and exhausted bulb. This is the method which, with some modifications, is 
employed by the author. 
Davy found that slow combustion goes on at temperatures considerably below 
that at which the mixture takes fire. The existence of a discontinuity between this 
gradual combustion and ignition proper is assumed by the phrase ‘temperature of 
ignition.’ The author shows that such a discontinuity does really exist in some 
eases, while in others, especially in mixtures containing a large proportion of 
an inert gas, there is a perfect gradation from the slow combination through a 
- combination lasting many seconds and attended only by a faint glow up to prac- 
tically instantaneous combination accompanied by a bright flame. The discon- 
tinuity is explained as due to the effect of the heat produced by the slow combustion 
_ of the mixture in raising the temperature of the gases above that of the bulb into 
_ which they have been introduced. In accordance with this explanation the tem- 
_ perature of ignition is higher in a narrow tube than in one of larger diameter. 
With most mixtures ignition takes place at temperatures not much above the 
minimum temperature only after the lapse of a period of delay which may amount 
_ to thirty seconds or more. During this interval it is supposed that the temperature 
of the mixture is being raised above that of the tube by slowcombination. At the 
same time that the temperature of the mixture is raised its composition is also 
altered by the slow combustion, and these two effects oppose one another. In some 
mixtures this change of composition brings about the extinction of the combustion 
unless it be rapid at the commencement, and then the phenomenon of a delay in the 
ignition is not observed. 
The slow combustion of CS, is perceptible at 100°, and is comparatively rapid 
at 130°. It is attended by the production of a reddish-brown solid, which is 
deposited on the sides of the tubes, and also issues as a finely-divided smoke with 
the gaseous products of combustion. This powder contains both carbon and sulphur, 
but its composition has not yet been thoroughly made out. SO, is produced in 
abundance, but very little, if any, CO, or CO. 
The temperature at which the slow combustion develops into ignition varies 
considerably with the composition of the mixture. It is lowest for mixtures con- 
taining a large excess of oxygen, and is raised distinctly by the addition of nitrogen 
or carbon dioxide, but much more by addition of sulphur dioxide. This will be 
seen from the following table giving the temperatures of ignition in a tube of 5mm. 
internal diameter :— 
Ree MOO Yip iss oc. «<> >. get ante ee 
Po nome ane. AG! SOA cite 
Pees 4 ee ee ee 
a ene EGO! 28 |, sf Yea aire 
ereriressO. 0 eee ge 
The effect of change of pressure on the ignition was also examined, and found to 
be somewhat complex. The general effect of rarefaction is to lessen the discon- 
tinuity in the phenomena, while raising the temperature, of ignition; and this is 
readily explained as due to the smaller frequency of the molecular encounters at a 
low pressure. On the other hand, the extinguishing power of SO, was found to be 
-much diminished by rarefaction, the mixture CS,+50,+5SO, igniting at 195° 
ander a pressure of 150mm. ; and this has an influence on the ignition of mixtures 
which contained no SO, originally, since that gas is formed during the slow com- 
_ bustion in the period of delay which precedes ignition, Thus the mixture CS, + 100, 
_ under a pressure of 750 mm. ignites at 160° after a delay of 1-2 seconds, and under 
_ a pressure of 300mm. at 155° after a delay of as much as 15 seconds. These two 
effects of rarefaction act in opposite directions, and in some mixtures the one and in 
other mixtures the other of them has the preponderance. 
1890. 35 
