90 Mr. D. Waldie's Experimental Researches 



very luminous white portion surmounting the dark portion 

 which penetrates further up its centre than its circumference ; 

 this consists of solid particles of charcoal burning at a white 

 heat; 3rdly, a blue cap arising from the bottom and sur- 

 rounding the dark portion, extending moreover some way up 

 the sides of the white part, this being supposed to be com- 

 posed of the gas burning in an undecomposed state; 4thly, an 

 envelope of a light blue colour, of a pinkish or lilac shade, con- 

 sisting of the proper combustible mixture, this being the 

 hottest part of the flame : this part is thinnest at the bottom, 

 increasing in thickness to the top where, of course, the heat 

 is most intense; and, .5thly, another envelope of a yellowish 

 brown colour, chiefly covering the upper part, and consisting 

 of unconsumed matter or products of combustion. These 

 two latter are not well seen in the flame of coal gas, on ac- 

 count of the great size and brightness of the luminous white 

 portion. 



The explanation of these appearances is obvious enough : 

 the gas issuing from the jet spreads out into the atmosphere, 

 till it mixes with a sufficient quantity of oxygen to form a 

 combustible mixture; as it ascends, however, it becomes 

 strongly heated by the surrounding flame, deposits solid 

 charcoal, the combustion of which forms the white part of 

 the flame. 



When a jet of coal gas or defiant gas burns in oxygen the 

 parts of the flame are all the same, but differing considerably 

 in appearance ; the dark portion, the white portion, and the 

 blue cap 1, b, «, (Plate I.) fig. 1, are greatly diminished in 

 size, looking as it were compressed, the white being how- 

 ever more brilliant. The light blue portion again, 2, is 

 greatly enlarged, the shell of flame becoming much thicker, 

 and the heat is much greater, easily fusing a small platinum 

 wire. The yellow tail, 3, also is now perfectly visible. The 

 whole flame is also much smaller than in air. These changes 

 are due simply to the more perfect combustion produced by 

 the oxygen being undiluted, so that the greater part of the 

 gas at once undergoes perfect combustion; the diminution of 

 size of this as well as of flames in general, in oxygen com- 

 pared with air, arising from the circumstance, that in the latter 

 the gas requires to penetrate or diffuse itself over a much 

 larger space before it meets with a sufficient quantity of air 

 to produce full combination. 



In flame again, where there is no solid matter deposited, as 

 in that of hydrogen, the appearances are much simpler ; in this 

 case (burning in oxygen) it consists of a dark central por- 

 tion of unmixed gas, 1, of a light lilac blue envelope of ex- 



