304 SECTIONAL TRANSACTIONS.—B. 
of inflammable gaseous mixtures contained in a closed vessel. These experiments 
show that the introduction of infra-red radiation into the reacting system speeds 
up combustion provided 
(i) that the radiation introduced is of the kind which can be absorbed by the 
combustible gas, and 
(ii) that nitrogen is present in the gaseous mixtures. 
Thus the rate of combustion of mixtures of CO and oxygen and nitrogen is speeded 
up by 4-44 radiation (which is absorbed by CO), while that of mixtures of CH, and 
oxygen and nitrogen is speeded up by 3-2u radiation (which is absorbed by CHy,). 
Hydrogen has no absorption bands, and it is found that the introduction of radiation 
does not affect the rate of combustion of mixtures of H, and air. No speeding up of 
combustion is found in any of these mixtures when the nitrogen is replaced by oxygen, 
argon, or COs. 
Perhaps the most remarkable result obtained from the experiments is that 
different types of radiation produce the speeding up in the CH, and CO mixtures. 
Thus the effect of radiation is not due to the inhibiting of formation of oxides of 
nitrogen, for if this were the case, the same type of radiation should be effective, 
whatever the nature of the combustible gas. Rather the experiments suggest that 
some kind of temporary association between nitrogen and the combustible gas 
normally takes place during explosion, which association tends to be inhibited when 
the combustible gas is stimulated by radiation. 
Dr. O. C. pe C. Exiis.—Notes on the Production of Flame in Closed 
Vessels. 
When ignition of a combustible mixture occurs in a closed vessel the rate at 
which the flame begins to spread is compounded of the rate at which the ignition is 
communicated from layer to layer of the mixture, and the rate at which the mixture 
just about to be burnt moves under the expansive force of the burning mixture. 
The rate at which the flame begins to spread away from the source of ignition is 
therefore but little different whether the vessel be virtually without walls (as with a 
soap-bubble) or closed, for the initial combustion takes place at constant pressure. 
This state of affairs persists in a closed vessel for only a very short time; then the 
existence of the enclosure manifests itself. 
The first influence of the enclosure is on the direction of expansion of the burning 
gas, which is toward the more open space. If the enclosure has an opening in it the 
mean movement of the flame is toward that opening, and the flame will not travel 
through the remoter, completely enclosed part of the vessel until its other boundary 
has passed through the opening. In a strictly closed vessel the flame spreads from 
the region of ignition as if to reach all parts of the boundary at the same time. This 
is only a strong tendency, however, and is not actually accomplished except when igni- 
tion is central and the vessel spherical. It does not occur even then if the rate of 
spread of flame is so slow that convection currents can influence it. In such circum- 
stances the flame-front travels upwards more quickly than downwards. : 
The second effect of the enclosure, very marked in small vessels, is the production 
of a second phase of slower propagation, following the first rapid spread of the flame. 
In the first phase we have propagation assisted, in respect of spatial displacement, by 
expansion ; in the second, it is probable that propagation takes place against a slight 
recoil. In the first phase the gas burning is of low density ; in the second phase 
the gas is burning under higher and increasing pressure. It is the second phase 
which is indicated in a gas explosion by the fairly straight portion of the ascending 
line of the time-pressure curve. 
The third effect of the enclosure, most marked in rich mixtures, takes place among 
the flame-products, and is noticeable as a glow, occupying the space where originally 
the flame passed during its first phase ; that is to say, it begins near the point of igni- 
tion. It begins, moreover, well before the flame has completed its passage throughout 
the vessel, so that it is not an effect of cooling. A dark zone lies between the flame- 
front and this glow at the core, suggesting that the glowing material comprises the 
products of combustion of the first phase and the dark zone those of the second. 
As soon as the flame reaches the boundary of the vessel the dark zone must cool 
and therefore contract. For this reason, presumably, the glow now expands to fill 
the entire vessel. It may, and usually does, persist for a period many times as great 
as that necessary for the complete spread of the flame through the sphere. During 
this post-flame period its volume gradually grows less and its mass-centre rises. 
