BULLETIN OF THE BUSSEY INSTITUTION. 103 
afterwards with phenylhydrazin acetate, with the usual precau- 
tions, these mannite crystals gave, beside osazones, character- 
istic balls of mannose-hydrazone, which, on being recrystallized 
from 3:1 alcohol, gave excellent crystals and plates of mannose- 
hydrazone. 
Sometimes, as in the case of fleshy roots, like the turnip, per- 
colation is impracticable because the soft flesh is impermeable to 
water. In this event the material under examination was washed 
by way of decantation, using repeated small quantities of boiling 
water. 
Tests for Mannite in Various Grasses. 
As an example of the practical working of the process of perco- 
lation the following record of an experiment on oat hay may be 
cited. This hay was cut at the Bussey farm at the end of June, 
1902, when the grain was ‘‘ in the milk,” and was stored in a dry 
barn until November, 1903. It was cut into small pieces and 
ground in a drug-mill to a fine meal. 100 grm. of the powder, 
packed in the percolator, were leached with boiling water until 
some 150 to 200 c.c. of percolate were obtained. After evapo- 
ration to a volume of 100 ¢c.c., some of the ammoniacal copper 
sulphate was added. ‘The concentrated percolate was so dark- 
colored that it was not easy to see whether any precipitate fell; 
but, on filtering, a small quantity of a light green precipitate was 
collected. After washing with cold water this precipitate was 
decomposed with sulphuretted hydrogen, and the filtrate from the 
copper sulphide was evaporated to a small bulk and allowed to 
cool slowly. <A yellowish residue was obtained, which under the 
microscope showed crystals of mannite. When tested with ferrous 
sulphate and hydrogen dioxide, followed by phenylhydrazin, large 
balls of mannose-hydrazone were obtained, which yielded fairly 
good crystals and plates of this substance on being recrystallized 
from 3:1 alcohol. Jt should here be said that in an earlier experi- 
ment with the oat hay, where basic lead acetate, bone black, and 
alcohol were employed, no satisfactory evidence of the presence 
of mannite was obtained. 
In the case of a sample of June grass (Poa pratensis) hay 
mown in June, 1903, and stored in a dry barn until November, 
both crystals of mannite and thereafter mannose hydrazone crystals 
were obtained, as they had been from the oat hay, by the method 
