452 BULLETIN OF THE BUSSEY INSTITUTION. 
of course, upon aqueous solutions after the alcohol had been 
driven off, —the references are to dextrose. It is evident that 
the matter insoluble in alcohol contains some substance of higher 
rotatory power than dextrose, while as regards the portion soluble 
in alcohol it is to be inferred that it contained much xylose, the 
true specific rotation of which is (a) D = 19°.2. Calculated as 
dextrose, the matter insoluble in alcohol amounted to 4.65% of 
the dry birch wood, and the sugar soluble in alcohol to 12.54%. 
The second half of the filtrate, from the three hours’ boiling 
with hydrochloric acid of 83%, was now boiled three hours longer 
and neutralized and decolorized as before. It gave quasi 
(a) D = 40°.54 and ‘* dextrose” amounting to 9.97% of the 
dry wood, showing that in the absence of wood some of the pro- 
ducts of the first. hydrolysis were readily destroyed by the hot 
dilute acid. 
On evaporating this solution to dryness and treating the residue 
with cold, strong alcohol, the matter less soluble in alcohol gave 
quasi(a) D = 42°.27, and the readily soluble portion gave quasi 
(a) D = 227.172 , 
The residual cellulose from the wood which had been boiled 
with 3% hydrochloric acid was washed and allowed to dry in the 
air; it was then left to soak for 24 hours in 100 c¢.c. of hydro- 
chloric acid of 835%, and the mixture was poured into enough 
boiling water to reduce the acid to the strength of 3%. The 
mixture was boiled during three hours over a free flame, as before. 
Evidently the strong hydrochloric acid had but little action upon 
the wood. 
One half of the filtrate was neutralized with sodium hydroxide, 
evaporated and decolorized. It gave quasi(a) D = 85°.7 and 
dextrose equal to 0.98% of the dry residue, or to 0.78% of the 
dry wood. It is to be remarked, however, that this observation 
is of little or no value, because the solution was extremely dilute. 
It contained absolutely too little sugar to admit of a satisfactory — 
examination. Nevertheless, it was evaporated to dryness and 
treated with cold, strong alcohol. ‘The portion soluble in alcohol 
gave quasi(a) D = 53°.34. 
The second half of the filtrate, obtained after soaking in strong 
hydrochloric acid and subsequently boiling in this acid diluted to 
3%, was boiled three hours longer, but so little sugar was con- 
