M. R. Bunsen's Blowpipe Experiments. 515 



The temperature of the flame, calculated from its composition, 

 gives the following results :— 2369° C. ; 2252° C. ; 2391° C. ; 

 2386° C. The differences are comparatively inconsiderable ; and 

 the temperature may be taken at 2300° C. But this temperature 

 is not everywhere the same, and is only obtained where the gas 

 meets with just the right quantity of air necessary for its com- 

 bustion. This point may be ascertained by the following experi- 

 ment, which is an application of the principle of Bunsen's pho- 

 tometer. A small partially stearinized photometric diaphragm 

 of drawing-paper is placed as a shade on one of the holders c c, 

 fig. 3; and the end of the platinum wire, bent at right angles 

 about 4 millims. from the end, is made incandescent in the 

 flame. By this the diaphragm becomes so strongly illuminated, 

 that the part not stearinized appears dark on a white ground on 

 the side away from the lamp. If this side be also illuminated, 

 and with a light as constant and as distant as possible, so that 

 the dark part upon the light ground just disappears^ and if the 

 wire, which is to be kept ignited to the same length, be moved to 

 another part of the flame, it will be readily seen whether the 

 tem])erature of this place is greater or less than that first ob- 

 served ; for in the former case the spot on the diaphragm will 

 appear dark on a light ground, and in the latter light on a dark 

 ground. By trying in this manner the different parts of the 

 flame, it will be fouud that the hottest part of the flame is in a 

 zone of the external mantle a d a c, which extends a few milli- 

 metres above and below the diameter of the flame at b. This 

 region, which may be called the melting -space, is used to deter- 

 mine the deportment of bodies at the temperature 2300° C. The 

 external edge of this melting-space acts as an oxidizing flame, and 

 the internal as a reducing flame. The latter is most powerful just 

 below the point /;. 



The great constancy of the flame permits it to be used for de- 

 termining the volatility of bodies at 2300° C. This is effected 

 by melting on the platinum wire as much of the substance to be 

 tested as will form, when seen under the microscope, a bead of 

 1 millim. in diameter: with a little practice this is very easy. The 

 bead is then introduced into the flame, and the seconds whi(-h 

 2 J. 2 



