Theory of Gunpowder. 



509 



expand, the temperature of the flame will be difiierent. We 

 obtain this temperature by dividing the heat of combustion of 

 powder by the specific heat for a constant volume. The latter 

 is obtained by a calculation similar to the above. 



ax. 



Sulphate of potash . . 0*4554 



Carbonate of potash . 0-1362 



Sulphide of potassium . 0*0229 



Nitre 0-0401 



Charcoal 0-0079 



Sulphur 00015 



Nitrogen 0-1075 



Carbonic acid . . . 0-2167 



Carbonic oxide . . . 0-0101 



Hydrogen 0-0002 



Oxygen 0-0015 



Temperature of the flame 



619-5 

 0-18547 



0-18547 



=3340° C. 



Hence the temperature of the flame of powder whose gases are 

 formed in a closed space and cannot freely expand is 3340° C. 



If the flame consisted only of ignited gaseous substances, 

 whose capacity for heat, as has been shown by Regnault's ex- 

 periments and Clausius's theoretical considerations, does not 



change with the temperature, the value of - might be deter- 



s ° 

 mined with great accuracy. But since the specific heat of solid 

 bodies increases with the temperature, the numbers 2993° and 

 3340° can only be accepted as approximate values, which cannot, 

 however, from what we know of the magnitude of the increase 

 with the temperature, be very far from the truth. Since s in- 

 creases with the temperature, - would be found too great; and 



since, in fact, the temperature of the flame is continually lowered 

 by radiation and conduction, we may assume that the tempera- 

 tures 3340° and 2993° represent limits to which the temperature 

 of the flame more or less approximates, but which it can never 

 fully reach or exceed. From this fact an important conclusion 

 may be drawn as to the amount of pressure exerted by the powder 

 in the space which it gravimetricaliy fills. 



It has been hitherto generally assumed, that during the com- 

 bustion of powder the residue assumes the vaporous condition, 

 and by its tension materially helps the mechanical effects of the 

 powder. Although a small volatilization cannot be denied, it 



