Experimental and Applied 253 



the evidence that we have of a species of saline or ethereal con- 

 stitution which characterises cellulose in many of its reactions. It 

 is probable that the OH groups of opposite function exert a 

 mutually repressive influence ; and acetic anhydride, being essen- 

 tially a condensing agent, is unable to determine the liberation of 

 the CH groups into the condition in which they can react. 



On the other hand, the presence of certain reagents in relatively 

 small proportion is sufficient to disturb the equilibrium, and 

 reaction results. An elevated temperature also determines reac- 

 tion, and obviously for a similar reason, viz. that the equilibrium 

 of the molecule holds only for a certain range of conditions. 



For these reasons it may be considered doubtful whether a 

 cellulose acetate in the strictest sense can be obtained. Acetates 

 are, however, obtainable which satisfy the { general ' a priori defini- 

 tion viz. which by saponification yield a carbohydrate having the 

 negative characteristics of cellulose, and in contradistinction to 

 acetates which, though obtained from cellulose, yield on saponifi- 

 cation a carbohydrate more or less soluble in the alkaline solution, 

 and reducing CuO. 



Parchmentising Process. The student should make careful 

 statistical observations upon this reaction. An unsized paper 

 should be used ordinary * waterleaf ' or filtering paper. The acid 

 should be placed in a * photographic ' dish of suitable size ; the 

 sheet immersed in this and, after suitable exposure to the action 

 of the acid, transferred at once to a dish of water. After staying 

 in this first wash water for some time they must be thoroughly 

 washed in a flow of water until neutral. The sheets having been 

 weighed before treatment, and the moisture estimated in portion 

 of the same paper the product must also be weighed after drying. 

 It is necessary also to determine the cellulose dissolved by the 

 process, for which purpose the acid and first wash water may be 

 treated. These may be mixed and a certain fraction boiled, the 

 dextrose formed being estimated by any of the well-known methods ; 

 or the solution may be neutralised with lime evaporated, and the 

 total carbon estimated, by the method of combustion, with CrO s 

 and H 2 SO 4 . 



The product may also be further examined (i) for loss of 

 weight in boiling with dilute alkaline solutions ; (2) CuO reduction 

 in presence of alkalis (Fehling's solution) ; and (3) degree of 



