TRANSACTIONS OF THE SECTIONS. 69. 
pitate, redissolve and throw down a second time. This precaution is necessary, as the 
first product is contaminated with oxalate of magnesia. The lime precipitate con- 
tains generally a variable quantity of phosphates of iron and alumina. To estimate 
these, dissolve it in hydrochloric acid, add ammonia, wash precipitate, redissolve from 
the jilter, and throw down a second time without heat. When washed and dried, as 
burned, the weight is deducted from the first amount of which it is a part, and the re- 
sult is pure carbonate of Jime. The phosphate of iron and alumina being dissolved in 
hydrochloric acid, a little tartaric acid is added, and the phosphoric acid finally deter- 
mined as phosphate of magnesia, 
On the Carbonates of Alumina, Chromic Owide, and Ferric Oxide. 
By W. Watrace, Ph.D., F.CS. 
On Chloro-Arsenious Acid, and some of its Compounds. 
By Witt1aAmM Watrace, Ph.D., F.CS. 
In the course of a former series of experiments: on: the chloride of arsenic*, I 
observed that arsenious acid dissolved freely in the-anhydrous chloride. Believing 
that a definite combination was formed, I have recently-investigated the subject more 
fully, and have succeeded in preparing a new and -highly interesting compound, to 
which I have given the name of chloro-arsenious acid. My examination of the acid 
and its compounds is still far from complete ; but I purpose in the mean time to sub- 
mit a short abstract of my results to the members of the British Association. 
1, Solution of Arsenious Acid in Chloride of Arsenic.—When chloride of arsenic 
is heated to gentle ebullition in a small tubulated retort, the beak of which is inclined 
upwards, and powdered arsenious acid is gradually introduced, the solution of the lat- 
ter appears to cease when the liquid contains equal equivalents of the two compounds. 
The readiest method of preparing this solution in quantity is to introduce a few ounces 
of arsenious acid into a flask; and pass hydrochloric acid gas through it until all the 
arsenious acid disappears. ‘The flask should be agitated occasionally, The action is 
very violent, and is attended by the elimination of much heat. If the passage of the 
hydrochloric acid gas is continued as long as absorption takes place, pure chloride of 
arsenic is obtained. The nature of the reaction has been fully explained in the paper 
already referred to. 
2. Chloro-arsenious Acid.—When the solution of arsenious acid in chloride of 
arsenic is gradually distilled until it begins to froth up, there separates, on cooling, a 
pasty, viscid, semi-fluid mass, from which the more liquid portion may be poured off. 
Analysis gave as follows :— 
ATEONAG SKS Gili bie <5 1=/5 59°29 
Chieriney oss aieattons: 241 =35:5 28-06 
ORVEON gs oh medicwr iy (ay 2=16 12-65 
. 126°5 100-00 
The formula of this compound, therefore, is AsCl?4+2AsO* or AsClO2, that is, 
arsenious acid in which one equivalent of oxygen is replaced by chlorine. 
Anhydrous chloro-arsenious acid is a viscid fluid or a very soft solid, according to 
the temperature to which it is exposed. It is transparent, but has a brown colour, 
which does not appear to be owing to the presence of any impurity. It fumes slightly - 
in the air, parting with a small portion of its chlorine as hydrochloric acid, and 
absorbing oxygen. When strongly heated, it boils up with considerable frothing, and 
affords a distillate of pure chloride of arsenic. When brought up to about the tempe- 
rature at which arsenious acid sublimes, it leaves a glassy, hard, transparent substance, 
which was found to contain 10°94 per cent. of chlorine, agreeing with the formula 
2AsO°, AsCl0?. 
The fluid poured off from the chloro-arsenious acid was found, on analysis, to con- 
tain an amount of chlorine which corresponds with ‘the formula AsClI°, AsO*. It is 
therefore similar to the solution prepared by adding arsenious acid to heated chloride 
of arsenic. I do not believe, however, that this liquid is a compound of these two 
* “ On Chloride of Arsenic,” by Penny and Wallace, Phil. Mag. vol. iv. p. 861. 
