ON VALERIANATE OF IRON. 
133 
twelve hours longer, but with no further loss. In order to ascertain whether 
any acid would be lost by artificial heat, it was then transferred to a water- 
bath, and kept at a temperature of 212° until the weight remained constant; 
the loss was then 3*41 grains, or very nearly 25 per cent., but no change had 
taken place in the appearance of the substance, and no apparent loss of acid 
by smell or otherwise. It had now lost all stickiness and become quite firm, 
so that it might readily be chipped oflT the vessel in thin transparent scales, 
like the usual citrate or tartrate of iron. A few of these scales were treated 
with alcohol, and were found to dissolve readily to a clear deep ruby liquid, 
but they were quite insoluble in water, and if water is added to the alcoholic 
solution the valerianate is immediately precipitated. 
A quantity of the dry scales, weighing 9’1 grains, were then gently heated 
over a gas lamp, when they immediately began to melt, and give ofi’ copious 
fumes of valerianic acid, which readily ignited and gave a yellow brilliant 
flame. The residue, which appeared to consist of oxide of iron, weighed 2‘31 
grains ; this was dissolved in a little hydrochloric acid, and the oxide of iron 
precipitated with ammonia. The weight of precipitate so obtained was 2*03 
grains ; the loss of weight between this and the ignited residue was due to 
sulphate of soda not thoroughly washed out of the valerianate. 
I had no time to ascertain the quantity of valerianic acid actually present 
in the salt ; but assuming that the loss from ignition was due to this acid 
alone, the formula of the salt would be as near as possible Fe 2 03+3 Va, or a 
neutral anhydrous valerianate of peroxide of iron, containing three equivalents 
of valerianic acid=:279, and one equivalent of oxide of iron = 80; and this is 
the formula given by Wittstein, but which he says he was not able to pro¬ 
duce, his explanation being as follows :—“ If a neutral solution of valerianate 
of soda is added to one of perchloride or any persalt of iron, a dark brick-red 
precipitate of valerianate of iron is formed, together with a sulphate or chlo¬ 
ride of sodium, according to whichever iron salt has been used. In the case 
of perchloride of iron— 
3 atoms NaO +Va and 1 atom Fe 2 C ]3 
form 
1 atom Fe 203 -t- 3 Va and 3 atoms NaCl. 
The neutral compound of valerianic acid described above does not, however, 
precipitate,” he goes on to say, “but decomposes as soon as formed into a 
basic salt, in which, if the formula of the neutral salt is tripled, two atoms or 
acid are replaced by two atoms of water, forming a salt 3Fe2 03 -j- 7 Va + 2H0 
and free valerianic acid, which, holding a portion of the basic salt in solution, 
gives to the supernatant liquor a slightly yellowish-red colour.” This basic 
salt is therefore the substance generally found in commerce, and I suppose is 
the only valerianate of iron which it has been hitherto thought could be ob¬ 
tained. The neutral valerianate is in ruby scales, will dissolve in its own weight 
of alcohol, and will bear heating to a temperature of 212° without decomposi¬ 
tion. I7ot only in these respects but in appearance it differs completely from 
the brick-red powder hitherto known to us, and I doubt not will be found a 
much more reliable article for medicinal use. 
The next question for our consideration is the preparation of the salt, and 
herein seems to lie the secret of the difierence between one salt and the 
other. My experiments in this matter have not been so extensive as I could 
wish, but so far as they go they seem to show that the use of dilute solutions 
of valerianate of soda and iron necessitate the production of the basic salt, and 
vice versa, with concentrated solutions. Another thing is of great impor¬ 
tance, namely, that the valerianate of soda used should be entirely free from 
carbonate or caustic soda, a thing of rare occurrence with the usual com¬ 
mercial article apparently, more especially the carbonate ; this may be known 
