288 On Sir H. Davy's ScJ'c.-lamp, mid on Flame. 



losopliv, is made' void bv my experiments with parallel rods not 

 intersecting to form meshes. I am happy to afford you a cor- 

 rect solution, the inference of actual experiment. In the results 

 of the experiments detailed in a former number of the Philoso- 

 phical Magazine, I have stated my belief tl\at the pha?nomena 

 were neither electrical nor magnetic. I find that if the rod in- 

 troduced into the flame of a candle approximate the wick, it is 

 invested with a cloud of aquemis vapour, which fills vp the 

 chasm to which "I have alluded. This emanation of aqueous 

 vapour will suffer condensation by contact with the meshes of 

 Mire-gauze cooled bv the ambient atmosphere ; flame cannot 

 therefore run this gauntlet— for, in fact, the meshes are filled up 

 by a transparent screen of aqueous vapour. The cause of the 

 interval between the surface of metal to which I have in my for- 

 mer paper referred, is now conipletelv and satisfactorily solved. 



I have made some experiments with flame — the aqueous par- 

 ticles seem chiefly disengaged from the sides of the cone of 

 flame. The upi)cr i)art of the spire evolves the charcoal and 

 heated gases. Sir H. Davy observes, that if burning phosphorus 

 is introduced into a large extent of flaming alcohol, its flame 

 will be seen within that of the other : but I observe, if the base 

 of the flaming cone is from one to two inches diameter, that 

 both sulphur and phosphorus, when plunged into it, previously 

 ignited, are severally instantly extinguished; nor does potassium 

 even burn in the interior, 1 have supposed that probably, when 

 a larger surface is employed, the uneven undulating surface might 

 occasion chasms, through which the exterior air might enter and 

 support the flame : indeed, the following experiment is analogi- 

 cally in favour of the opinion. I adopted a value opening up- 

 wards into the cone, and when the alcohol was burning, little 

 j)uffs of air broke through the lateral surface of the cone. 

 I am, with high respect, sir, 



Your very humble servant, 

 Stranraer, N. B, Oct. 10, 181C. J. MuRIlAV. 



P. S. — Respecting the ignition of gunpowder by flame, I sim- 

 ply alluded to the introduction of it on a slip of ivory into the 

 cone. The cone of flame is not a solid or uniform mass, but is 

 hollow within: — this is proved bv pressing the apex by a plate 

 of glass, and looking down through it ; the mantle of flame sur- 

 rounding the wick will thus appear not amch thicker than a 

 wafer. J. M. 



LVIII. Ahs- 



