which is preferable to diffuse daylight for making this observation.—For the success of the oxidation test it is indispensable that the inside of the tube be not soiled with an alkaline flux: because the alkali would unite with the oxide. It is also requisite not to heat the tube suddenly to redness before the oxide is sublimed; because then the oxide is apt to unite with the glass, forming a white, opaque enamel. The physical characters of the sublimed oxide are so delicate and precise, that they may be accurately distinguished, even when those of the metallic crust are obscure, owing to its minuteness. Sometimes too, the metal may be so scanty that it is oxidated at once in the act of subliming, and never presents the appearance of a metallic crust. Although the characters of the crystalline oxide in either of these cases are very precise and distinctive, it may be right to subject it to a farther test when the metal is not previously exhibited with its characteristic properties. For this purpose it is sufficient to cut away with a file the portion of the tube which contains the sublimate, to boil it in another tube with a few drops of distilled water till the sublimate disappear, and then to test the solution with one of the fluid tests to be presently described, the ammoniacal nitrate of silver.
After all that has been recently written as to the old and newer processes for detecting arsenic, I must nevertheless avow my conviction, that for solid arsenic no test is, for medico-legal purposes, at once so satisfactory, convenient, and delicate as the test of reduction, especially with the addition of the supplementary test of oxidation. That other methods are still more delicate may be readily granted. But where the suspected substance is in the solid form, what possible occasion can there be for a method more delicate than one which will detect a 300th part of a grain? A method ten times less so would meet every case in actual practice.—A variety of supplementary tests have been proposed. But they are all greatly inferior in facility, or conclusiveness, or both, to the process of oxidation, and ought therefore to be expelled from medico-legal practice,—not even excepting the alliaceous odour of metallic arsenic in the act of subliming, a character, the fallaciousness of which was long ago pointed out by myself as well as others, and to which a preposterous importance has been attached in some late inquiries. The reader will find in the last edition of this work an attempt to estimate the value of various tests supplementary to that of reduction. This disquisition is now omitted, as it seems no longer necessary.
Of the Tests for Oxide of Arsenic in Solution.
Oxide of arsenic in a state of solution may be detected in one of four ways; by what are called the liquid tests; by precipitating it with one of these, and subliming metallic arsenic from the precipitate, which method is usually termed the reduction process; by Marsh's method, which consists in disengaging it in the form of arseniuretted-hydrogen gas, and decomposing the gas by combustion; or by the
