BULLETIN OF THE BUSSEY INSTITUTION. 59 
Roots of Dahlia. From gardens at the Bussey Institution and at Hingham, 
Mass. 
Sweet Potato (Jpomea batatas). Bought in Boston market in May, 1903. 
Only the flesh proper was examined, after removal of the skin. 
Roots, 2.e. root-stalks of Sweet-flag (Acorus calamus). Dug up not far 
from the Bussey Institution, May 20, 1903. 
Rootlets of Sweet-flag, cut off from beneath the root stalks of the fore- 
going paragraph. 
With regard to the method of testing, in all of the foregoing 
experiments, it will be noted, in those cases where only negative 
results were obtained, that it is possible that positive results 
might have been got if quantities of the materials larger than 
20 grams had been operated upon. The comparatively small 
quantity (20 grams) was chosen and adhered to because the plan 
of the research was merely to ascertain in a general way what 
kinds of plants carry an abundance of mannan, and at what sea- 
sons of the year, as distinguished from plants that contain little 
or none of it. 
Mannose Forms in Presence of Alkalies. 
Another point to be emphasized is a certain risk that some 
mannose might possibly be formed from the dextrose in the pro- 
ducts of the hydrolyses unless extreme care were exercised in the 
process of neutralizing these products by means of sodium hydrox- 
ide, as a preliminary to their evaporation. Lobry de Bruyn* has 
shown that dextrose when subjected to the action of alkalies is 
readily changed in some part to mannose, and since dextrose 
must necessarily be present in the products of the hydrolysis of 
wood which contains cellulose, it would manifestly be improper 
to leave any free alkali in contact with these products of hydro- 
lysis during the process of concentrating them by evaporation. 
If any free alkali were present some mannose would be formed 
from the dextrose, and would be exhibited as a matter of course 
on the addition of phenylhydrazin acetate. 
In the series of tests of woods, etc., recorded in this article, 
special pains were taken to avoid making the products of hydro- 
lysis actually alkaline by means of sodium hydroxide, and to have 
the liquids neutral or even very slightly acid, in some cases, at 
* Recueil des Travaux Chimiques des Pays Bas, 1895, 14, 156; and Lobry 
de Bruyn and Ekenstein, zbid. 14. 203, 18. 147. 
