58 Transadtions of the 



the actual existence of an elemental fire, maintained that it was 

 but the result of the commotion of the particles of the heated 

 body. This doctrine descended to a late period of European 

 philosophy, and formed a part of the scientific creed of Bacon, 

 Boyle, Descartes, and Newton, and in still later days has been 

 revived in another form, "Heat a mode of motion." Ideas on 

 the actual constitution of flame were always vague in the extreme. 

 Modern science has, however, we have every reason to believe, 

 successfully solved the difficulty. We consider flame as the par- 

 ticles of a gas or vapour, raised to a state of incandescence by 

 the heat evolved from the combination of the burning body with 

 oxygen or other supporter of the combustion. Flames freqiiently 

 contain solid particles, the temporary or periiianent result of the 

 chemical changes which are taking place. It is on the presence 

 of these particles that the luminosity of a flame depends, for the 

 presence of solid particles in the flame raises its brightness by 

 acting as radiant points of light. 



The lecturer now proceeded to illustrate this portion of his 

 paper. Pure hydrogen was shown to emit a feeble light when 

 burnt in air, but on the introduction of such bodies as platinum, 

 charcoal, &c., which could increase the radiant power of the flame, 

 or on saturating the gas with benzole, it became luminous. Simi- 

 larly phosphorus and iron, the products of whose combustion in 

 oxygen are solids, burnt in that gas with a brilliant light ; but 

 the combustion of phosphorus in chlorine, the product being 

 gaseous, was of feeble luminosity. Similar experiments were then 

 tried upon an alcohol flame, with platinum, asbestos, paper soaked 

 in calcic chloride, oxide of zinc, &c. ' Compounds containing carbon, 

 from which a portion of the carbon is separated in the form of soot 

 by the heat of combustion, are therefore luminous, as marsh-gas, 

 olefiant-gas, ether-oils, fats, resins, &c. It is noticeable that the 

 complete combustion of ether (whose flame is luminous) requires 

 twice the amount of oxygen requisite for alcohol (whose flame 

 is not luminous) — 



(a HO HO + 30, = 2C0, -I- 3H,0 

 (C, Hs),0 + 60, =: 4C0, + 5H,0. 

 It appears probable that, in ordinary air, all the oxygen required 

 for the combustion of ether cannot be readily obtained, and that 

 the unoxidised particles of the carbon act as radiant points of 

 light.' 



The non-luminous alcohol flame may be rendered luminous, by 

 combining part of the hydrogen with chlorine, and thus freeing 

 carbon ; and the luminous coal gas flame may be rendered non- 

 luminous by mixing it with air or oxygen, and thus completing 

 tlie combustion of the carbon, as in the Bunsen gas-lamp. The 



