on Combustion and Flame. 89 



really but also apparently as much combustibles as hydrogen 

 or coal gas, and these again are exhibited in the form of sup- 

 porters of combustion. In fact this distinction arises simply 

 from the accidental circumstance of oxygen being contained 

 in our atmosphere, and of hydro-carbonaceous substances 

 being emitted from other bodies into it: there is really no 

 distinction between the two, the phaenomena of combustion 

 proceeding from the act of combination of the two oppositely 

 electrical substances. 



I did not succeed in making iodine vapour burn in hydro- 

 gen, nor the reverse ; the affinity for each other seems not 

 sufficiently strong. Nitric oxide does not burn in hydrogen, 

 nor vice versa ; a circumstance remarkable enough, as the two 

 constituents of this gas are in respect to their density in pre - 

 cisely similar circumstances to a simple mixture of the two ; 

 no more however than might have been expected from the 

 circumstance that a mixture of the two does not detonate but 

 burns with flame in contact with the air. It is said in some 

 works on chemistry, that the products of this combustion are 

 water and pure nitrogen : this I suppose must have been a 

 conclusion made beforehand, on the supposition that the nitric 

 oxide is decomposed completely: it would appear rather that 

 the oxygen for the combustion of the hydrogen is supplied 

 from the air, and that the nitric oxide is either not decom- 

 posed, or at least only partially, as I find copious nitrous acid 

 fumes produced, whether the experiment is performed by 

 burning them together in a jar even with excess of hydrogen, 

 or by burning the mixture from a jet either in oxygen, or oxy- 

 gen in it. 



I had made but very few of these experiments until I per- 

 ceived that there was a great variety in the appearance of the 

 flames. The jet of hydrogen for instance, in passing from air 

 to oxygen was observed to shrink in size, and become brighter 

 and denser ; the jet of oxygen in hydrogen was observed 

 also to be much smaller than that of hydrogen in oxygen. 

 The flame of hydrogen in chlorine, nitrous oxide, or nitrous 

 acid was much larger than in air ; the flame of these gases in 

 hydrogen again was small and concentrated. In all these 

 cases the gases were made to issue from the same jet, with 

 as nearly as possible equal degrees of velocity. 



Before making any observations on the causes of these dif- 

 ferences, it will be necessary to take some notice of the struc- 

 ture of flame. This has been studied chiefly in the flame of 

 a common candle or jet of coal gas. It consists then, accord- 

 ing to observations already made, of, 1st, a dark central por- 

 tion consisting of unmixed, unconsumed gas ; 2ndly, a dense 



