1 88 Cellulose 



numbers corresponding approximately with the empirical 

 formula C 6 H 10 O. ; , which is confirmed by more recent results 

 (Tollens). The only value of these results, however, is to esta- 

 blish a normal carbohydrate ' formula. The more important 

 problem of its constitution has been elucidated, as already noted, 

 by its yielding the pentaglucose xylose as the main product of 

 proximate hydrolysis (acid), and furfural as the ultimate product 

 (HC1). The most recent analyses of Tollens gave the following 

 results : 



Furfural Xylose 

 /Specimen i . . .3878 74-26 



Wood gum J .!! ' ' ' 46-90 89-82 



,, iii . . 48-08 92-02 



* iv . . . 33-30 6373 



These specimens were from beech wood, variously pre- 

 pared : No. i by the process already described ; Nos. ii and 

 iii by extraction with alkaline solution, after boiling the raw 

 material with dilute sulphuric acid ; No. iv by extraction with 

 boiling milk of lime. It is evident from these results that 

 wood gum is a pentosan the amorphous anhydro- aggregate 

 of xylose or xylan mixed or combined with variable propor- 

 tions of a carbohydrate of similar empirical composition, prob- 

 ably a cellulose derivative. It also generally contains meth- 

 oxyl (2 '6 p.ct. O.CH 3 ). These observations further confirm the 

 view that the pentosans are derived from hexose groups, and 

 represent the final terms of a series of transformations of which 

 * wood gum ' as directly obtained may be taken as representing 

 the intermediate terms. 



METHOXYL DETERMINATIONS. The O.CH 3 group, as a 

 chemical constant of lig?rification, has been brought into promi- 

 nence by the investigations of Benedikt and Bamberger ; their 

 most important communication on the subject appearing 

 under the title, * Ueber eine quantitative Reaction des Lignins ' 

 (Monatsh. n, 260-267). Employing the perfected method of 



