U. S. DEPARTMENT OF AGRICULTURE. 
elm, wild cherry, linden and willow, having an equal content of tannin, for feeding 
the animals during the winter, and no inconvenience or accident results. 
The experiments, partly with tannin, partly with tannic acid made by the pharma- 
cologists, no more prove the harmfulness of these substances or the production of 
hemoglobinuria. Gohier has performed in this line an experiment often quoted. 
He fed large quantities of oak bark to horses, which came out with good results; he 
proved an arrest of digestion with obstinate constipation, but without hematuria. 
Their blood, far from having a tendency to lose its hemoglobin, became more red, 
more coagulable, and much slower in putrefying. 
Moreover the therapeutic investigations made with tannic acid employed as pure 
as possible have not disclosed hematuria following its use for a long time. 
Finally we may add that the consumption of the leaves of oak, ash, etc., in summer, 
autumn, and winter at a time when they contain a much smaller proportion of tannin 
than in the spring, it is true, but still in large quantity, does not bring on the "maladie 
des bois." In the mountain region of the southeast and center, especially in the high 
and low Alps, the Lozere, the Ardeche, the Haute-Loire, the Rhone, etc., they cut in 
autumn the leafy branches of the oak growing from the pollards, and with these they 
feed the sheep for a long time without accident. 
If the tannin of the bark, fruit, and leaves of summer and autumn can not be 
blamed, two hypotheses are presented; there may exist in the young leaves along with 
the tannin an ephemeral poisonous substance, which soon disappears, or the tannin 
itself may come into a special condition which may give it the harmful properties 
which have been described. 
The first hypothesis is hardly tenable for it would be very singular if this poison 
should have, up to the present time, escaped the notice of the chemists and botanists, 
who have studied tannin so thoroughly. We can not deny it absolutely, for we can not 
pledge the future, and can not foresee what may ultimately be discovered, but we 
expect the facts will support this belief. 
Let us look at the second. It is admitted nowadays that plants hold the t annin 
under the form of a polygallic glucoside which is easily changed. The varieties of 
this are many, following the plant species which furnish them. It is probable that 
in the same species many of these varieties, derived one from the other, appear and 
disappear to make way finally for the specific variety. It will be very desirable that 
the chemists should study this point, commencing their analyses with the opening of 
spring. The check of its plant histology should not be neglected. We possess, 
actually, many good reagents which can furnish useful information: the perchlorid 
which colors green or blue, according to the character of the tannin; the bichromate of 
potassium which forms a compact reddish-brown compound and the solution of molyb- 
date of ammonia in a concentrated solution of chlorhydrate of ammonia, which colors 
the tannin red and has the advantage of making it possible to distinguish the glucosides 
of tannin from the tannic acid, for an excess of the chlorhydrate of ammonia produces 
in the first voluminous precipitate, while the last remains colored red. 
These varieties of tannin, very changeable as has been said, may suffer modifications 
in the plant economy and furnish sometimes acids, sometimes special bodies like 
pyrocatechin. 
Now, some very interesting investigations of M. C. Hayem have shown that the 
blood of animals submitted to the action of pyrogallic acid and pyrocatechin shows 
special modifications. The blood corpuscles are attacked and a certain proportion 
of hemoglobin extravasated. There is a formation of methemoglobin at the same 
time in the red corpuscles and in the plasma and more or less intense degeneration of 
the corpuscles. On the other side it has been established that in hematuric fever and 
in paroxysmal hemoglobinuria, the urine contains methemoglobin. 
