subsequent numbness indicate _aconite_; intense bitterness points to _strychnia_; if there be no taste at all it is unlikely that any alkaloid is present. There are some alkaloids of a peppery taste; these are irritants, and are not common as poisons. Bitterness is the most frequent characteristic. 2. Moisten the contents of the second watch glass with a little water and a trace of acetic acid, and apply through an incision in the skin of the back of a young frog. He should be kept as comfortable as possible and the symptoms observed. Strychnia readily produces tetanus in this animal; other poisons also have peculiar effects. Some observers have used mice, rabbits, or cats; in the Palmer trial it was observed that dogs were not employed because they were inconvenient and might bite! On the whole this so-called physiological test has been overrated, as it is hardly to be expected that an animal with its back cut and otherwise injured will not exhibit some symptoms; and all who have kept wild animals in confinement will know how soon they become, first almost convulsive from excitement, then finally sink into stupor and die. If necessary, any judge may grant a special licence to the experts in a trial to make experiments on animals, otherwise such cruelty is rendered penal by the Vivisection Act.[3] 3. To the third watch glass, after the contents have been dissolved as before, a drop of a solution of iodine in potassium iodide is added. Nearly all alkaloids give a brown precipitate. If none occur, a negative conclusion may be expected. 4. Test the fourth watch glass in one corner for strychnia by concentrated sulphuric acid and peroxide of manganese; in another corner for morphia by iodic acid and starch; in a third corner for brucia (and morphia) by strong nitric acid. (See the special paragraphs on these reactions, pp. 280, 285.) 5. If there is still no indication, and no information has been obtained from other sources, it may be necessary to employ Dragendorff’s process on the remainder. But if the poison has been discovered, the solution reserved in the burette should be evaporated, dissolved in water and a little dilute acid, avoiding heat, and titrated by Mayer’s reagent to ascertain the quantity.[4] The second and third extractions of the organs must now be considered. Most of the text-books recommend that _all the extracts should be mixed_. The objection to this is, that since the alkaloid is usually present in very small amount, the first extraction will remove nearly all of it, while the second and third will mainly contain other matters, and therefore will be only adding to the impurities, and consequently to the difficulty of isolation. If it be worth while, the second and third extracts may be treated _separately_ as above, and should any further quantity of alkaloid be found, it may be determined quantitatively, and the amount added to that already obtained.