B.- -CHEMISTRY. 37 



A. 'J'rimetliyl nielhylglucoside. 



B. Trimethyl inethylglucoside. 



C. Trimethyl luetiiylglucosido containing ;i small proportion of 



dimetliyl methylglucoside. 

 All the fractions were analysed, and in this way it was shown that 

 the small quantity of dimethyl methylglucoside in fraction C agreed 

 exactly with the deficiency of 2.6 per cent, in the methoxyl content 

 of the trimethyl cellulose used. No trace of tetraviethyl methylglucoside 

 was present. Moreover, the physical constants of the trimethyl methyl- 

 glucoside agreed exactly with those recently established for this com- 

 pound by Irvine and Hirst." On hydrolysis of fractions A and B an 

 89 per cent, yield of crystalline 2,3,6-trimethyl glucose was obtained. 

 The identity of this product was confirmed by analysis, by mixed 

 melting-point with an authentic specimen, and by the mutarotation in 

 aqueous solution 



[a]„-fl08° > + 67.0°. 



No isomeric trimethyl glucose was present; higher and lower methyl- 

 ated glucoses were absent. We thus reach the conclusion that tri- 

 methyl cellulose gives 2,3,6-trimethyl glucose as the only product. 

 The reactions involved in the research are shown below, and, consider- 

 ing the nature of the operations involved, the yields may be claimed to 

 be quantitative. 



Cellulose 

 I 



I 

 Trimethyl cellulose , 



! Yield 90";: 



2,3,6-Trimethyl methylglucoside 



I I Yield 89% 



2,3,6-Trimethyl glucose ' 



The scheme affords a proof that all the glucose residues ina-cellulose 

 are identical in structure, and the simplest possible formula which will 

 satisfy this condition is that of a 1,5-anhydro-glucose. 



I ° 1 



CH-CHOH-CHOH-CH-CHCH.,OH VI. 



I ^ 



It is necessary, however, to include at least one additional glucose unit 

 to account for the formation of cellobiose,^ and this is fulfilled by the 

 formula 



-CH O 







CHOH 



CHOH 



I 

 -CH 



CH CH • CHOH • CHOH ■ CH ■ CH • CH.OH 



CH2OH O 



VII. 



