124 ' i Cellulose 



of too indefinite a character to require more than a passing 

 notice. Generally we may regard reaction as occurring only 

 with such as undergo pronounced dissociation on solution in 

 water, which dissociation is exaggerated by the fibre-substance. 

 The lignocelluloses present merely a particular case of the 

 general theory of the action of * mordants,' its combinations 

 with the 'mordanting' oxides being similar to those of the 

 cellulose, differing only in the higher proportions of the oxides 

 taken up. 



There is one reaction, however, of a specific character, 

 already alluded to, which merits description in detail that is, 

 the interaction of the lignocellulose and ferric ferricyanide. 



This reaction has been described in a paper by the authors 

 in the J. Soc. Chem. Ind. 1893, the experimental portion of 

 which is now reproduced, with certain alterations of minor 

 import. 



A REACTION OF THE LIGNOCELLULOSES AND THE THEORY 

 OF DYEING. The red solution obtained by mixing aqueous 

 solutions of ferric chloride and an alkaline ferricyanide, which 

 may be regarded as containing ferric ferricyanide, reacts as is 

 well known with the more easily oxidisable ' organic ' com- 

 pounds, oxidising them and being itself reduced to the lower 

 mixed cyanides, i.e. Prussian blue and similar compounds 

 (Watts' Diet. ii. 248). The reactions of this solution with 

 the lignocelluloses, and notably jute, are remarkable and cha- 

 racteristic. Not only is the conversion into the blue pigment 

 very considerable in proportion to the weight of fibre- substance, 

 but the colouring matter is deposited within the fibre in such 

 a way as to give the effect of a homogeneous dye. Thus if 

 this particular fibre, suitably purified by previously boiling in 

 dilute alkaline solutions and washing, be plunged into a \ deci- 

 normal solution of the reagent (prepared by mixing decinorrnal 

 solutions of the reagents in equal volumes) the fibre-substance 



