248 ANNUAL OF SCIENTIFIC DISCOVERY. 



Hydrate of potash 1-26 



Sulphocyanid of potassium 0-86 



Saltpetre 5-19 



Carbon 0-97 



Sulphur trace. 



100-52 

 The condensed smoke, or powder- vapor, was found to contain 



Sulphate of potash 65-29 



Carbouate of potash 23-48 



Hyposulphite of potash 4'90 



Hydrate of potash 1-33 



Sulphocyanid of potassium 0-55 



Nitrate of potash < 2-48 



Carbon 1-86 



2-3 Carbonate of ammonia. . . 0-11 



100-00 



These analyses show clearly that the residues of the explosion of gunpow- 

 der consist essentially of sulphate and carbonate of potash, and not of sulphid 

 of potassium, as is erroneously assumed in military books. The gaseous 

 products of the explosion were found to be 



Carbonic acid 52-G7 



Nitrogen 41-12 



Carbonic oxide 3-88 



Hydrogen 1-21 



Sulphid of hydrogen 0-GO 



Oxygen 0-52 



100-00 



From this analysis it also appears that the old theory of the explosion is 

 incorrect, since, according to it, nitrogen and carbonic acid should be present 

 in the ratio of 1:3; whereas the actual ratio is not even as 1 : 1*5. The au- 

 thors found, further, that one gram of the powder used by them gave 0*6806 

 gr. of solid residue, and 193*1 c. c. of gas, or one-third less than should be 

 given according to the old theory. To determine theoretically the work done 

 by the explosion, the authors first determined the quantity of heat evolved 

 in the combustion, which was found to be in heat units G43'9 C. for one gram 

 of powder. This number requires, hoAvever, to be corrected for the heat 

 evolved by the combustion of the gaseous products in the combustion-tube 

 filled with air, amounting to 24'4 units, so that the true heat of combustion 

 is 619*5 C. The calculated quantity of heat, supposing that the combustible 

 ingredients of the powder burn freely in oxygen, is 1039*1 C. : the difference 

 is easily explained by the absorption of heat on the part of nitrogen set free. 

 The temperature of the flame of the powder is found by dividing the num- 

 ber 619*5 by the specific heat of the products of the explosion, namely 0*206, 

 and is therefore equal to 2993 C. When, however, powder burns in a space 

 in which it cannot expand, the number 619*5 must be divided by the specific 

 heat of the products of the combustion under a constant volume, which is 

 0-18-547, and the temperature of the flame is therefore 3340 C. These values, 

 2993 C. and 3340 C., are approximations which cannot be far from the 

 truth. The authors, in the next place, show that the received theory that the 



