152 On the Combustion of compressed Gas. 
tion, and without loss of arsenic. The melted salt is then to 
be dissolved in water, and lime added in excess, and the liquid 
boiled. The arseniate of lime will then be deposited, and 
may be collected. When dried, it is mixed with charcoal, 
and then brought to a red heat by the blowpipe, and a small 
quantity of this mixture is allowed to fall to the narrow end of 
the above tube. It is now gradually heated to expel all humi- 
dity which tends to throw it into the wide tube, and when it is 
very dry, heat at the flame of the blowpipe, the part of the 
tube-which contains the mixture. The arsenic will be disen- 
gaged, and be sublimed, at a distance from the heated part. 
An addition of vitrified boracic acid greatly promotes the de- 
composition which then takes place at a less elevated tempe- 
rature; but this acid frequently contains water, and produces 
a bubblirig of the melted matter which raises it in the tube, 
and causes the vapours to issue by perforating the softened 
part of the ia , : 
M. Berzelius maintains, that the sixth part of a grain of sul- 
phuret of arsenic is sufficient to make three different trials; but 
he adds, that, when we have discovered only very small traces 
of arsenic, we must take care not to introduce any by means 
of re-agents, among which both the sulphuric and the hydro- 
ghloric acid may contain it. The first almost always contains 
some arsenic when it is not manufactured from voleanic sul- 
phur ; and the second, in consequence of sulphuric acid being 
used in the preparation of the hydrochloric acid, yields the 
arsenic which it contains in separating it from soda. We 
must, therefore, be certain of the purity of these re-agents. _ 
When death has been caused by the arsenic, and not by the 
arsenious acid, the process must be modified, because the sul- 
phuretted hydrogen gas decomposes the arsenic:acid too slowly. 
In this case, we must add hydrosulphuret of ammonia, which 
reduces the arsenic acid to the state of sulphuret, which is af- 
terwards precipitated by the hydrochloric acid.— Edin. Journ. 
of Science. ; 
ON THE COMBUSTION OF COMPRESSED GAS.—BY MR. DAVIES. 
In. making, upwards of twelve months ago, some experi- 
ments upon the combustion of compressed gas, I accidentally 
observed a fact which is, I think, of rather a singular nature. 
When the aperture of the burner is, in this ease, too large, 
the flame cannot be maintained, being blown away bythe rapid 
current of the gas. When it is rather small, the flame is under 
the best circumstances. If the aperture be further enlarged 
without being carried to the extent at which the combustion is 
extinguished, the flame will then be blue, noisy, and agitated, 
affording very little light. But I found, to my great surprise, 
‘ that 
