A wooden,--or better, a gutta-percha--cylinder (Fig. 1), 5 cubic centimetres in height and from 20 to 25 c. c. in diameter, is employed. A piece of moistened parchment is securely attached to one of the openings of the cylinder, which, upon drying, shrinks and completely closes the aperture. If its continuity becomes impaired, the pores of the membrane should be covered with the white of an egg which is subsequently coagulated by the application of heat. The organs previously cut into small pieces, or the materials found in the alimentary canal, etc., after having been allowed to digest for 24 hours in water at 32°[A]--or, in dilute acids, if the presence of an alkaloid is suspected,--are then placed in the upper vessel, which is termed the dialyser. The whole should form a layer not over 2 cubic centimetres in height. The dialyser is next placed in the larger vessel filled with distilled water. In about 24 hours three-quarters of the crystalloid substances present will have passed into the lower vessel. The solution is then evaporated over a water-bath, and submitted to analysis. The portion remaining in the dialyser is decomposed by one of the methods previously described, in order to effect the detection of any poisonous substances possibly present. Instead of the above apparatus, the one represented in Fig. 2 can be employed. The fluid under examination is placed in a bell-shaped jar, open at the top and closed below with a piece of parchment, which is then suspended in the centre of a larger vessel containing water. In other respects the operation is performed in the same manner as with the apparatus represented in Fig. 1. [Illustration: Fig. 2.] [A] The degrees of temperature given in the text refer to the centigrade Thermometer; their equivalents on the Fahrenheit scale can be obtained by means of the formula: 9/5 C° + 32 = F°. --_Trans._ II. DETECTION OF POISONS, THE PRESENCE OF WHICH IS SUSPECTED. DETECTION OF ARSENIC. It is frequently required, in chemical jurisprudence, to institute a search for arsenic in the remains of a deceased person, whose death is supposed to have been caused by the reception of a poison. Under these circumstances the poison is mixed with a mass of substances which would obscure its characteristic properties, and it becomes necessary, in order to accomplish its identification, to isolate it, and then, by decisive reactions, determine its character. Three methods exist which permit of this result; they are: 1st. The method used prior to Marsh's test. 2nd. Marsh's test. 3rd. A method more recent than Marsh's, proposed by _M. Raspail_. METHOD USED PRIOR TO MARSH'S TEST.