24 BULLETIN OF THE BUSSEY INSTITUTION. 
the clear filtrate, the yellow amorphous precipitate again appeared, 
and after a time crystalline balls (‘‘ sea-urchins ”’) of dextrose- 
osazone were observed, although there was no special evidence 
that the yellow dust and the osazone crystals were in any way 
connected with one another. Similar results were obtained in 
other experiments where dilute solutions of dextrose were treated 
methodically with successive portions of phenylhydrazin com- 
pounds in the hope of discovering the nature of the yellowish 
amorphous precipitate, above mentioned, which forms when com- 
pounds of phenylhydrazin are added to dilute solutions of dex- 
trose. Under the microscope this precipitate seems at first to be 
merely a fine amorphous dust from which no ea. have been 
obtained on treating with warm diluted alcohol and subsequently 
cooling and evaporating. But after the mixture has stood for 
several days, patches of a thin, greenish membrane, like the skin 
of a delicate fruit, are found admixed with the yellow dust, and 
after a while small crystalline masses of dextrose-osazone become 
visible, usually in the shape of little greenish balls thickly beset 
with fine, sharp-pointed spines, though sometimes bundles. of 
needle-like crystals were noticed or bunches of needles radiating 
from a central point. The small, green, spiny balls are specially 
apt to form in or upon the green, skin-like patches. As was just 
now said, it does not appear that the formation of the spiny balls 
is dependent in any way on the presence of the yellow, amorphous 
precipitate; they come manifestly from the oxidation of the dex- 
trose-hydrazone which has been present in the solution from the 
moment when phenylhydrazin was added to the dextrose. As 
contrasted with crystals of mannose-hydrazone, the crystals of 
dextrose-osazone formed under the conditions of this particular 
experiment are much more delicate. They consist usually of 
slender needles, often springing from a central point; even the 
spiny balls, above mentioned, seem to be really masses of needles 
closely impacted one against the others. 
For these trials, solutions were prepared by dissolving either 
0.2 grm. of pure dextrose (from Merck & Co.) in 10 or 5 ce. of 
distilled water, or 0.02 grm. of this dextrose in 10 ec. of water, 
and adding to 1 cc. portions of these solutions either 1, 2, 3 or 8 
drops of solutions of phenylhydrazin compounds, prepared as 
described on page 14. After a few minutes the liquids thus 
