540 MM. Pelouze and Maurcy on the New Methods 



venience. As regards the silicate of soda, it would have intro- 

 duced an uncertainty into the determination of the atomic for- 

 mula of gun-cotton. 



The third series comprises three specimens made by the 

 Bouchet method : one kept from the old manufacture of 1847, 

 and two others made recently. For one of these latter the Lenk 

 mixture was used (1 of nitric to 3 of sulphuric acid), and for 

 the other a mixture of 1 volume of nitric and 2 volumes of sul- 

 phuric acid ; which is equal to 1 part by weight of the former to 

 2 "46 of the second acid. This proportion was designated at 

 Bouchet by the name of unequal volumes. 



II. Quantity of Gun-cotton yielded by a given iveight of cellulose. 



Disregarding the few thousandths of foreign matters, purified 

 cotton is cellulose, C 12 H 10 O 10 , or C 24 H 20 O 20 . In a German 

 Report signed by MM. Redtenbacher, Schrotter, and Schneider, 

 the following formula is assigned to gun-cotton, 



C 12 H 7 7 , 3N0 5 , or C ,2 H 7 (N0 4 ) 3 10 , 



which corresponds to the following composition : — 



Carbon 24'24 



Hydrogen 2*36 



Oxygen 59"26 



Nitrogen 14*14 



100-00 



The equation of the reaction may be stated in two ways : — 



1. By assuming that, in contact with the mixture of nitric 

 and sulphuric acids, the cotton loses water, which is replaced by 

 the first of these acids, 



C 12 H 10 O 10 + 3NO 5 =C 12 H 7 O 7 3NO 5 + 3HO. 



2. By supposing that the hydrogen of the cellulose is replaced 

 by an equal number of equivalents of hyponitrous acid, 



CJia H 10 O l0 + 3NO 5 =C 12 H 7 (NO 4 ) 3 O 10 + 3HO. 



According to this, 100 parts of cotton ought to give 183 of gun- 

 cotton. But by varying in more than 100 experiments the 

 proportions of the bodies concerned, we could never obtain more 

 than 178 parts. 



The German Report is silent regarding the yield, which, in 

 our opinion, is the most solid basis for the composition of gun- 

 cotton. We do not mean to say thereby that the exact determi- 

 nation of the quantity of gun-cotton produced renders useless 

 the elementary analysis of the latter ; but it is necessary that the 

 analysis agree with the quantity of the product. 



Our experiments on the yield were made with cotton of good 



