38 
THE TANNINS. 
water and precipitated it with one part of isinglass, 
washed, dried, weighed the precipitate, and calculated 
40 per cent, of it as tannin. 
Muller suggested the addition of alum to the isin¬ 
glass or gelatin solution, and a solution of three 
grammes gelatin and one gramme alum in one litre of 
water is largely used at the present time in two modi¬ 
fications,—one to collect, wash, dry, weigh the precipi¬ 
tate, and reckon 54 per cent, of it as tannin ; the other 
to use the solution, volumetrically and determine the 
end of the reaction by filtering off a small portion of 
the solution and testing until no further precipitate is 
caused by the reagent. The solution of gelatin and 
alum must first be standardized by a solution of pure 
tannin. Notwithstauding the objection to this method, 
and the fact that we have many others for which 
greater accuracy is claimed, it continues to be used, and, 
no doubt, in careful and experienced hands will yield 
uniform results. 
The following details of this process have been em¬ 
ployed by the author for several years with consider¬ 
able success. The results have in most cases agreed 
closely with those obtained by the permanganate 
method. The solution is made of 2.5 grammes gelatin 
and 10 grammes alum in one litre. This solution, as 
well as that of the tannin, is warmed to 70°. The re¬ 
agent is then dropped in with constant stirring until 
the precipitate coagulates and leaves a clear superna¬ 
tant licpiid which gives no further precipitate on the 
addition of a few drops of the reagent. In case the 
coagulation does not take place so as to furnish a clear 
liquid and one that filters easily, the whole should be 
