ON THE ESTIMATION OF TANNIC ACID. 223 
Three samples of Jalapx resina , B.P., were examined with the following 
results:— 
Sol. in ether. Sol. in alcohoL Sol. in water. 
A. hard, but yielded on exposure to air 8*26 86'66 5‘08 
B. quite brittle, and retaining its 
consistency .... 
D. hard, but not brittle, speedily at-'I 
tracted moisture, forming a re- > 3751 44‘37 18T2 
mollient mass . . . . J 
Sample D. had evidently been prepared from Tampico jalap. 
ON THE ESTIMATION OF TANNIC ACID. 
BY JOHN WATTS, B.SC. 
In March, 1867, I published a paper in the ‘ Pharmaceutical Journal ’ upon 
the “ Various Processes which have been proposed for the estimation of 
Tannin,” in which I endeavoured to embody my own experience, and, at 
the same time, to test the relative efficiency of the several methods therein 
described. I wish now to supplement that paper by a few remarks, and to 
endeavour to show, if possible, the condition in which the subject remains 
down to the present date. 
Fleck's Process .—In the ‘ Bayerisches ICunst und Gewerbeblatt,’ 1860, pp. 
209-225, Dr. H. Fleck introduced a new method for the analysis of tanning 
materials, based upon the fact that both tannic and gallic acids are completely 
precipitated from solutions containing them by neutral cupric acetate, and, 
moreover, upon the property possessed by cupric gallate, to become redis¬ 
solved by a solution of ammonic carbonate. It is obvious, then, that the 
calculation of the percentage of tannic acid in any substance can be effected 
either by bringing the washed cupric tannate to the balance, or by the esti¬ 
mation of an excess of cupric salt. This method has, I understand, obtained 
considerable reputation in Continental laboratories, though, as far as I am 
aware, it has never appeared in any of our English chemical periodicals. The 
details of the process are as follows:— 
Standard Solutions.—A solution of neutral cupric acetate is prepared either 
directly from the commercial article, or by dissolving the diacetate (verdi¬ 
gris) in strong acetic acid, allowing the solution to crystallize, and redissolv¬ 
ing the crystals so obtained in thirty times their weight of distilled water. 
The solution is then ready for titration. 2 grammes of pure copper obtained 
by galvanic deposit are dissolved in nitric acid without heat, and diluted with 
pure water, so as to yield a volume bulk of 1 litre. Every c. c. of this solu¬ 
tion contains 2 milligrammes of copper. 10 c. c. of this liquid are then ruu 
into a beaker, slight excess of ammonic carbonate added, and a solution of 
cyanide of potassium dropped in from a burette until the blue colour disap¬ 
pears. The number of c. c. of cyanide required correspond to *002 grm. of 
pure copper. 
With care and practice this may be done with great exactitude, but to ob¬ 
tain consonant results it is necessary that the solutions of copper and am¬ 
monic carbonate should always be employed in the same relative proportion ; 
that is, if in one analysis 10 c. c. of copper be used in conjunction with 20 c. c. 
of ammonia, in a second experiment the same quantity of copper must not be 
employed -with 30 or 40 c. c. of ammonia, otherwise the proportion of cyanide 
required will not be constant. 
I 
3*08 
93-82 
3T0 
