252 Cellulose 



temperature : the reaction should proceed at the lowest limit 

 of temperature ; (2) the conditions of mercerisation : the reaction 

 would probably be favoured by employing a certain proportion of 

 zinc oxide in replacement of sodium hydrate ; and (3) the liquid 

 medium. There is, of course, a rapid hydrolysis of the benzoyl 

 chloride by the solution of sodium hydrate ; and there are various 

 ways which suggest themselves of retarding this waste reaction. 



The precipitation, as benzoates, of modifications of cellulose 

 soluble in alkaline solutions should prove a valuable aid to investi- 

 gation ; but, before the method can be applied to the investigation 

 of substances and mixtures of unknown or lesser known composi- 

 tion, it will be necessary to characterise the products obtainable 

 from the normal celluloses. There are in plant tissues a widely 

 diffused group of substances having the external characteristics of 

 the resistant celluloses, or celluloses proper, but which are readily 

 soluble in alkaline solutions. In the course of an ordinary 

 proximate analysis these are usually ' dismissed ' as ' non-nitro- 

 genous extractive matters ' ; which, of course, is but a very loose 

 description. 



We have seen that hydrated modifications of the normal cellu- 

 lose are themselves soluble in alkaline solution, and it would 

 appear that in this condition they are generally more reactive. 



There are two practical problems suggested by the properties 

 of these hydrates : (i) Do they yield to the processes of animal 

 digestion ? and (2) how far can the processes adopted by the 

 paper-maker for isolating cellulose from complex fibrous materials 

 be limited, to prevent the solution of celluloses either existing or 

 hydrated in the raw material or tending to pass into such modifi- 

 cations ? 



The investigation of these questions will be found to be greatly 

 facilitated by a method of separating the soluble celluloses from 

 solutions. The benzoates give promise of affording such a method 

 of separation, and invite investigation from this point of view as 

 much as from that of their intrinsic theoretical interest. 



Cellulose Acetates. The * suppression ' of the OH groups of 

 cellulose is evidenced by its not reacting with acetic anhydride 

 at its boiling point ; whereas the poly-hydroxy-compounds gene- 

 rally are acetylated under these conditions, and many even react 

 with acetic acid itself. We may find some explanation of this in 



