on Combustion and Flame. 95 



Gases were also mixed with nitrogen and carbonic acid. 

 When a pure gas was burnt in such a mixture it had its 

 flame much enlarged, as it had to expand itself further till it 

 met with a sufficient supply of the other, on account of its di- 

 lution. Thus hydrogen, as already noted, burns with a larger 

 flame in air than in oxygen, and oxygen with a very large 

 flame in hydrogen diluted with twice its bulk of nitrogen or 

 carbonic acid. When mixed with the gas which was to issue 

 from the jet an opposite effect was produced ; thus the mix- 

 ture of nitrogen or carbonic acid with hydrogen burnt with 

 a smaller flame in oxygen than pure hydrogen did, and air 

 burnt with a much smaller flame in hydrogen than oxygen did 

 — there being less gas issuing — and of course sooner meet- 

 ing with a sufficiency of hydrogen. In this case the flame of 

 air was such as represented in fig. 5, burning from a wide 

 hole in a piece of a tobacco pipe : 1, dark ', 2, blue very di- 

 stinct ; 3, dark reddish tail ; 4, greenish envelope. 



Carbonic acid was employed in these experiments, in order 

 that its effects might be compared with those of nitrogen, on 

 the supposition that, on account of its greater density, a jet of 

 gas burning in it would be expanded or diffused to a greater 

 degree than in a similar mixture of nitrogen; this accordingly 

 was found to be the case. The greater degree of contrac- 

 tion which should be supposed to take place when the mixed 

 gases were made to issue from the jet was not demonstrated 

 so clearly by the experiments ; in some instances it seemed 

 sufficiently apparent, in others it was doubtful. Experiments 

 with more accurate apparatus would be necessary to establish 

 the proportional effect of such mixtures. 



These experiments were made on the idea that from them 

 we should get an explanation of the fact, that carbonic acid 

 is more deleterious to combustion than nitrogen, or " exerts 

 a positive influence in checking combustion, as appears from 

 the fact, that a candle cannot burn in a gaseous mixture com- 

 posed of four measures of atmospheric air and one of carbonic 

 acid." (Turner's Elements, sixth edit.). On examining the 

 effect of nitrogen on the combustion of a jet of coal gas, it 

 was found that by successive additions of it to air the flame 

 became more and more expanded, the tube of flame becoming 

 wider and longer, and the white part diminishing or disap- 

 pearing altogether when the nitrogen was in considerable 

 excess; when beyond a certain proportion, it would not burn 

 at all, obviously from the cooling [effect produced by excess- 

 ive diffusion. It was therefore concluded, that this effect 

 depended on the density of the gases, and accordingly on 

 trial the following results were obtained. 



