GENERAL CHARACTERS. 
27 
is difficult to get rid of. This may be overcome by 
agitating the 5 cc. of glycerin directly with 20 cc. of 
ether without first dissolving in water, or the sample of 
tannin may be heated with 5 grammes of paraffin, in¬ 
stead of 5 cc. of glycerin, and the paraffin dissolved in 
petroleum ether; the residue is then treated with water 
and agitated with ether, the ethereal solution separated 
as before, evaporated, and the residue dissolved in 
50 cc. of water, when it is ready for testing. 
In order to determine the effect of dilute acids on a 
sample, a convenient quantity of the latter is heated for 
an hour or more to 100° with a two-per-cent, solution 
of absolute hydrochloric acid. This may be accom¬ 
plished by heating in a sealed tube or strong flask 
tightly stopped, or by attaching the open flask to an up¬ 
right condenser and heating with a vigorously boiling 
water-bath. The solution, after sufficient heating, is 
allowed to cool, when a separation of insoluble matter 
indicates ellagic acid, anhydrides, or phlobaphenes. 
The mixture is filtered, if a separation has taken place, 
and the precipitate tested for ellagic acid by nitric acid, 
with which it forms an intense crimson color. The 
filtrate containing hydrochloric acid, undecomposed 
tannin, gallic acid, and glucose is agitated with ether 
to remove gallic acid, neutralized with sodium hydrate, 
and precipitated with basic lead acetate to remove 
coloring-matter and tannin. The excess of lead is 
removed with sulphuric acid, again neutralized with 
sodium hydrate, and the clear filtrate heated with an 
excess of Feh ling’s solution. The formation of a red 
precipitate of cuprous oxide indicates the presence of 
glucose. 
