Compound Celluloses 1 21 



are discussed. Empirically the results of the treatment are as 

 follows : 



A certain proportion of the lignocellulose is dissolved ; but 

 the dissolved portion, as well as the fibrous residue, gives the 

 characteristic reactions of the original. When the latter is 

 chlorinated, and the cellulose isolated in the usual way, the 

 percentage yield is found to be unaffected by the treatment. 

 The character of the cellulose is somewhat altered, however, as 

 it is obtained in continuous strands ; and when dried, the 

 filaments of jute cellulose have a certain amount of coherence. 

 So far there is a general resemblance to the changes produced 

 in cotton cellulose on mercerising, i.e. the effects are chiefly 

 hydration changes. The differences, on the other hand, are 

 brought into evidence when the alkali-lignocellulose is exposed 

 to the action of carbon disulphide. 



The thiocarbonate reaction which ensues is of a remark- 

 able character. The lignocellulose is gelatinised more or less 

 in the reaction, but on treatment with water it is not dissolved 

 to a homogeneous solution, but swells up enormously, the 

 hydration proceeding to almost indefinite limits. The following 

 results of an experiment may be cited : 



4-5 grms. fibre, treated with excess of 12 p.ct. solution of 

 NaOH, squeezed, and exposed to CS 2 (2-0 grms.) in a stoppered 

 bottle 24 hours. On treatment with water, the gelatinised fibre 

 occupied a volume of 300 c.c. ; and for separation of the un- 

 dissolved fibre, dilution to 750 c.c. was necessary. The 

 following determinations were made : 



(a) Undissolved fibre 427 



(b) Dissolved reprecipitated by HC1 . . 43 *3 

 (b 1 ) Soluble after acidification . . . . .14*0 

 (a) Gave the reactions of the original fibre-substance. 



(b} Gave only a slight reaction with Cl and Na 2 SOs. 

 (b'} Consisted mainly of the furfural-yielding constituent. 



The following results of particular experiments are of interest : 



