explosive Mixtures confined Inj Wxre-gaitxe, ^c. 25 



inch, or 196 to the square inch, did not communicate explosion 

 till it became strongly red hot, when it was no longer safe in 

 explosive mixtures of coal gas ; but no motion tliat could i)e 

 given to it, by shaking it in a close jar, produced explosion. 



Iron wire-gauze of ^'^, and containing 240 apertures in the 

 square inch, was safe in explosive mixtures of coal gas, till it 

 became strongly red hot at the top. 



Iron wire-gauze of J-^, and of 24 apertures to the inch, or of 

 5 70 to the square inch, appeared safe under all circumstances 

 in explosive mixtures of coal gas. I kept up a continual flame 

 in a cylinder of this kind, eight inches high and two inches in 

 diameter, for a quarter of an hour, varying the proportions of 

 coal gas and air as far as was compatible with their inflamma- 

 tion; the top of the cylinder, for some minutes, was strongly red 

 hot ; but though the mixed gas was passed rapidly through it by 

 pressure from a gasometer and a pair of double bellows, so as 

 to make it a species of blast furnace, yet no explosion took 

 place. 



I mentioned in my last communication to the Society, that a 

 flame confined in a cylinder of very fine wire-gauze did not ex- 

 plode a mixture of oxygen and hydrogen, but that the gases 

 burnt in it with great vivacity. I have repeated this experiment 

 in nearly a jiint of the most explosive mixture of the two gases : 

 tlu'y burnt violently within the cylinder; but, though the upper 

 part became nearly white hot, yet no exj>losion was communi- 

 cated, and it was necessary to withdraw the cylinder to prevent 

 the brass wire from being melted. 



These results are best explained by considering the nature of 

 the flame of combustible bodies, which, in all cases, must be 

 considered as the combustion of an explosive mixture of inflam- 

 mable gas, or vapour and air; for it cannot be regarded as a 

 mere combustion at the surface of contact of the inflammable 

 matter: and the fact is proved by holding a taper or a piece of 

 burniiig phosphorus within a large flame made by the combustion 

 of alcohol, tiic flame of the candle or of the phosphorus will ap- 

 pear in the centre of the other flame, proving that there is oxy- 

 gen even in its interior part. 



The heat communicated by flame must depend upon its mass; 

 this is shown by the fact that the top of a slender cylinder of 

 wire-gauze hardly ever becomes dull red in the experiment on 

 an explosive mixture, whilst in a larger cylinder, made of the 

 same material, the central part of the top soon becomes bright 

 red. A large quantity of cold air thrown upon a small flame, 

 lowers its heat beyond the explosive point, and in extinguibbiug 

 a flame by blowing upon it, the effect is jnobably principally 



produced 



