522 Mr. Gladstone on the 



lyses recorded below it was cut into small pieces, and, after 

 the weight was taken, mixed carefully with oxide of copper. 

 To prevent its caking together the admixture of a little as- 

 bestos was found useful. This was introduced into a long 

 combustion-tube, then some fresh oxide of copper, and upon 

 it again some fused into lumps so as to fill the whole bore for 

 about 7 inches. Lastly, was added a mixture of copper turn- 

 ings and reduced copper for about 9 inches. The combus- 

 tion conducted cautiously in the usual manner gave the fol- 

 lowing results ; the pyroxyline burnt in the sixth experi- 

 ment having been prepared by Schbnbein's method. 



I. II. III. IV. V. VI. 



Pyroxyline employed 4-09 4-61 3*57 4-85 4-55 2-905 



Carb. acid produced 4*20 4-52 3*42 4*88 ... 2-84 



Water produced . . 1-19 1*36 1*34 0-87 



Hence in 100 parts, — 



I. II. III. IV. V. VI. 



Carbon . 27*90 26*74 26-10 27*44 ... 26-65 



Hydrogen 3*22 3*27 3*27 3*32 



In order to determine the amount of nitrogen the differ- 

 ential mode was adopted, as the method of MM. Will and 

 Varrentrapp is inapplicable to substances containing this 

 element in so highly oxidized a state. The same precautions 

 were taken as in the estimation of carbon ; and the collected 

 gases gave the following results after due correction for baro- 

 metrical pressure : — 



I. II. Another specimen. 



Carbonic acid . 25-0 38*5 23-9 



Nitrogen ... 5*5 8-5 5*1 



The volumes of the gases represent respectively equivalents 

 of carbon and nitrogen, and since no secondary product is 

 formed in the conversion of lignine into pyroxyline, the 24 

 equivalents of carbon in the former must be found in the 

 latter. This will give the following ratio in equivalents of 

 carbon and nitrogen according to the three experiments above 

 cited : — 



I. II. III. 



Carbon .... 24-0 24*0 24-0 

 Nitrogen . . . 5*28 5-3 5'12 



