If the death has really taken place from disease, and not from poison, or if it has been caused by poison, and yet no definite hint of the particular poison can be obtained either by the symptoms or by the attendant circumstances, the analyst has the difficult task of endeavouring to initiate a process of analysis which will be likely to discover any poison in the animal, vegetable, or mineral kingdom. For this purpose I have devised the following process, which differs from those that have hitherto been published mainly in the prominence given to operations in a high vacuum, and the utilisation of biological experiment as a matter of routine. Taking one of the most difficult cases that can occur--viz., one in which a small quantity only of an organic solid or fluid is available--the best method of procedure is the following:-- [Illustration] A small portion is reserved and examined microscopically, and, if thought desirable, submitted to various “cultivation” experiments. The greater portion is at once examined for volatile matters, and having been placed in a strong flask, and, if neutral or alkaline, feebly acidulated with tartaric acid, connected with a second or receiving flask by glass tubing and caoutchouc corks. The caoutchouc cork of the receiving flask has a double perforation, so as to be able, by a second bit of angle tubing, to be connected with the mercury-pump described in the author’s work on “Foods,” the figure of which is here repeated (see the accompanying figure). With a good water-pump having a sufficient length of fall-tube, a vacuum may be also obtained that for practical purposes is as efficient as one caused by mercury; if the fall-tube delivers outside the laboratory over a drain, no offensive odour is experienced when dealing with putrid, stinking liquids. A vacuum having been obtained, and the receiving-flask surrounded with ice, a distillate for preliminary testing may be generally got without the action of any external heat; but if this is too slow, the flask containing the substances or liquid under examination may be gently heated by a water-bath--water, volatile oils, a variety of volatile substances, such as prussic acid, hydrochloric acid, phosphorus, &c., if present, will distil over. It will be well to free in this way the substance, as much as possible, from volatile matters and water. When no more will come over, the distillate may be carefully examined by redistillation and the various appropriate tests. [Illustration: This figure is from “Foods.” B is a bell-jar, which can be adapted by a cork to a condenser; R is made of iron; the rim of the bell-jar is immersed in mercury, which the deep groove receives.]