Alcoholic Fluids, Oils, <^c. in Lamps. G3 



metal, bright sparks being observed ; but rods of different metals 

 produce similar effects ; and when numerous minute particles of 

 <;old or boiling water are made, by mechanical means, to impinge 

 on the exterior surface of the flame, the blue colour is not there- 

 by affected. A perfectly clean rod of glass, however, has the 

 same effect as rods of metal, only no sparks are observed ; and 

 hence particles of the water of wells, thus elicited, cause the ex- 

 trication of yellow light ; but pure water, that is, the pure com- 

 pound of hydrogen and oxygen, has not been procured for ex- 

 periment. When the particles of alcoholic fluids, or of water, 

 impinge on the interior or exterior surface of the flame, there is 

 doubtless an absorption of heat ; but the mere absorption of heat 

 cannot produce the observed eff^ects, as appears by an experi- 

 ment already noticed. If we approach a blue spirit flame, to 

 another of the same colour, no change is produced ; but, if a 

 flame of that colour be brought near to a yellow spirit flame, so 

 that the gaseous products of the latter may come into contact 

 with the former, the blue flame acquires a yellow colour. Hence 

 the products or substances emitted from a yellow flame are diffb- 

 rent from those of a blue flame ; and as steam, as formerly stated, 

 does not change the colour of a blue flame, we are led to trace 

 the extrication of yellow light to some other cause. 



It is known that carbonic oxide gas, in a certain state, and 

 likewise light hydro-carburet gas, give out yellow light during 

 their combustion. If a splinter of wood (and various other ve- 

 getable substances may be used) be lighted, and in a few se- 

 conds again extinguished, the white vapour or smoke that issues 

 from it gives a fine yellow colour to blue flame. If the carbo- 

 nized extremity of the wood be brought into contact, or only 

 near to the flame, there is a profuse extrication of yellow light 

 from the exterior or brush flame ; and if the extremity of the 

 carbonized wood be held quite above the flame, there is a co- 

 pious discharge of yellow light, similar to that of the brush 

 flame, but which might, with more accuracy, be termed lumi- 

 nous vapour than a flame. In all these cases, it is possible that 

 a minute quantity of aqueous vapour may be present ; but if, 

 after the wood has been lighted, and the flame extinguished, the 

 combustion of the carbonized portion be allowed to proceed, un- 

 til an attenuated, extremely light, and spider-web-like substance 

 alone remains, the minutest particle of this substance, when 



