Dr. J. H. Gladstone on the Violet Flame of many Chlorides. 419 



by side. Chloride of mercury in the gas flame from a Bunsen's 

 burner gives this light also. A hydrogen flame is capable 

 of producing it from the chlorides of nickel and cobalt ; but it 

 requires the heat of red-hot coals to obtain it from chemi- 

 cally pure chloride of sodium, potassium, or barium ; indeed in 

 the latter case the tire must be very intense, but there is then no 

 mistaking the characteristic violet. Chloride of zinc also exhi- 

 bited it at this high temperature, and so did chloride of iron, 

 though less distinctly ; but a doubt must rest on such observa- 

 tions made with a coal-fire, as it is quite conceivable that these 

 chlorides may give up their chlorine to the alkalies or earths of 

 the ash. Chloride of silver was doubtful ; and I did not succeed 

 in obtaining the violet colour from either chloride of calcium, 

 lead, or manganese. 



That this light should be emitted from chemically pure chlo- 

 ride of barium at a very high temperature has a peculiar interest, 

 since it was this chloride which first attracted Alexander Mitscher- 

 lich's notice by green lines not belonging to barium, that make 

 their appearance when it is ignited in association with chloride 

 of ammonium. These green lines do not coincide with those of 

 the violet flame. 



The question naturally arises, to what is this violet flame due ? 

 Is it the chloride itself in the gaseous form that emits these rays 

 when heated up to a certain temperature, which differs according 

 to the metal with which the chlorine is combined ? Or is the pecu- 

 liar flame produced by chlorine when the chloride is decomposed 

 by heat ? Or does it depend on the combination of the chlorine 

 with the carbon or hydrogen of the combustible ? The latter 

 supposition is negatived by the fact that anhydrous chloride of 

 copper emits these rays equally whether it be placed in a flame of 

 hydrogen or of pure bisulphide of carbon. It is difficult to accept 

 the second supposition ; for though chloride of copper or gold is 

 certainly decomposed in the spirit-lamp flame, chloride of nickel 

 or iron is so likewise, and chloride of mercury is reduced to the 

 subchloride, and yet these last do not exhibit the coloured flame 

 at that temperature. Besides, a stream of chlorine or of hydro- 

 chloric acid passed into a flame never gives the violet light ; nor 

 does Dutch liquid, muriatic ether, or chloroform mixed with 

 alcohol and burnt in a spirit-lamp. The rays in question also 

 appear to bear no relation to those emitted in Pliicker's experi- 

 ments. The origin of this violet light evidently, therefore, 

 requires further elucidation. 



I remain, Gentlemen, 



Yours faithfully, 



28 Pembridge Gardens, J. H. GLADSTONE. 



November 15, 1862. 



