612 ON DETECTING THE 
cannot be obtained upon a small bit, or thin piece, 
of glass, and the result of the experiment is such as 
would lead one to suppose that the materials are fit 
to be used in an investigation of poisoning ; but 
when a coid solid substance of the bulk of several 
cubic inches, as, for instance, the thick end of a 
Wedewood’s ware pestle, is held to the flame, a 
dense metallic crust may be obtained: the rea- 
son of this is easily explained,—the heat com- 
municated to a small bit, or thin piece, of glass, 
by the combustion of the hydrogen, soon becomes 
so great as to keep the arsenic in a volatile state; 
but a large solid substance is a long time in be- 
coming hot, and, consequently, upon it the metal 
continuously, though slowly, and by little and 
little, deposits, until ultimately a very distinct and 
even dense crust is obtained. It is obvious, that 
if this precaution of holding a large cold substance 
to the flame be not attended to in testing the purity 
of the materials, and if then, by the application of 
such large substance to the flame of gas produced 
after the introduction, into the apparatus, of any 
matter suspected to contain arsenic, a deposit of 
that metal be obtained, a very great risk is run of 
forming an erroneous and dangerous decision. 
In consequence of the sulphuric acid of com- 
e 
