272 
THE FUNCTION OF TANNIN. 
coloration of the contents of a cell is no proof that those 
cells in their uninjured state contained tannin. Further, 
all these salts, and especially the chloride, are uncertain. I 
used German preparations of “ Eisenchlorid,” or of “ Scliwe- 
felsaures Eisenoxydul,” * * * § in all of the observations hereafter 
detailed, as a preliminary, in order to test the nature of the 
tannin present. I took all the customary precautions, bat 
sometimes my notes were “ no tannin,” when the subsequent 
use of another salt, or of potassium bichromate showed it 
distributed widely, if not plentifully. I cannot, therefore, agree 
entirely with Nageli,f when, speaking of the reactions of 
the iron oxide salts, he says “ Diese Farben treten . 
in so intensiven Tonen hervor, dass sie der Jodreaction auf 
Starke an Sicherheit und Zulassigkeit kaum naclistehen.” 
Perhaps in another sense of the comparison I can however 
coincide, since I have already shown J that very frequently 
the iodine reaction for starch is unreliable, unless the reagent 
be decidedly aqueous. 
Sanio has recorded still another general objection. “ Gerb- 
saures Eisenoxya ist in iiberschussigen Eisenchlorid loslicli.” § 
Take a section of gall apple in water; add one drop of iron 
chloride and you have a dark-blue precipitate; add iron 
chloride to excess, the precipitate dissolves to a greenish 
yellow, or, if much precipitate be present, to an olive-green 
clear fluid. 
The same author had previously || recommended the use 
of dilute iodized chloride of zinc (Chlorzinc iodine, Schultz’s 
solution), by which tannin is coloured rose-red to violet. 
The same general objection of easy diffusibility applies equally 
to this reaction, with the additional one that the cell walls 
and any starch contents of the cells are alike coloured from 
red to violet, thus addiug an element of confusion, especially 
in the case of researches like the present, where the presence 
of starch in the same cells with the tannin is an important 
feature. 
* This latter I have found by far the more certain. 
f Ncigeli und Schicendener, “Das Mikroskop,” 2te Auflage, 1887, 
p. 490. Further on (p. 491), speaking of Sanio’s Potassium bichromate 
reaction, he says, “ Wir haben dieses Yerfahren selbst nicht gepriift (!), 
doch scheint uns die Beliandlung frischer Schnitte mit Eisenchlorid 
wenigstens fur saftige Gewebe den Vorzug zu verdienen.” This, after 
Sanio’s reaction had been in use for 14 years; and the preference, too, 
is given to the salt of iron which, of all those in general use, is the 
most uncertain. 
I “ Proceedings of the Camb. Phil. Soc.,” 1883, p. 400, 402. 
§ Sanio , “Einige Bemerkungen ii. den Gerbstoff ii. seine Yerbreitung 
bei den Holzpflanzen.” Bot. Zeit., 1863. 
I I Sanio, Bot. Zeitung, 1860, p. 213. 
