The difficulties abovementioned have led me to investigate the oxygen combining with platina by means of the nitrous gas yielded upon its solution in nitro-muriatic acid. By three distinct experiments I found that 10 grains of pure platina by solution yielded nearly 750 grain measures of nitrous gas, which may be considered as 1 grain in weight; ⅞ of which = .875 for oxygen; this would give 8.75 oxygen per cent. But from a subsequent experiment made under circumstances calculated to preclude as much as possible every source of fallacy, I obtained 790 measures of nitrous gas from 10 grains of platina; and the solution afterwards took 60 measures of oxymuriatic acid gas before a permanent smell of the gas was produced. These 790 measures = 1.05 grain, ⅞ of which = .92, to which add .04 for the oxygen acquired from the oxymuriatic acid, and we have .96 oxygen for 10 platina; or 100 platina take 9.6 oxygen. But if 9.6 ∶ 100 ∷ 7 ∶ 73, for the weight of an atom of platina, and 80 for that of the protoxide, as I apprehend it to be, and the only oxide of platina we can at present form. (The atom of platina in Vol. 1, page 248, was estimated at 100.) 3. _Oxide of Silver._ When silver wire is exploded by electricity in oxygen gas, a black powder is produced, which is the oxide of silver. If silver be dissolved in nitric acid and precipitated by lime water, an olive brown powder falls which becomes black when exposed to the light. This black powder is the only oxide of silver with which we are acquainted. The proportion of silver and oxygen has been investigated by various chemists; the results are as under. silver oxygen Wenzel 100 + 8.5 Proust + 9.5 Bucholz and Rose + 9.5[2] Gay Lussac + 7.6[3] Berzelius + 7.925 [Footnote 2: 7.9 when duly corrected. Annal. de Chimie, 78--114.] [Footnote 3: Memoirs d’Arcueil 2--168.] From the solution of 170 grains of standard silver I obtained nearly 30 oz. measures of nitrous gas = 18½ grains, corresponding to 16 oxygen. This would give 9.4 oxygen upon 100 silver. But as ⅒ of the metal or 17 grains was copper, and this takes ¼ of its weight of oxygen, we shall have 159 silver and 11¾ oxygen, or 100 silver and 7.7 oxygen nearly. If we adopt 7.8 as the proper quantity of oxygen on 100 silver, we shall have 7.8 ∶ 100 ∷ 7 ∶ 90 nearly, which represents the weight of an atom of silver, and 97 that of the oxide. 4. _Oxides of Mercury._ Two oxides of mercury have been long known and are well distinguished from each other. They may be obtained by exposing mercury to a heat not exceeding 600°, in contact with oxygen gas or atmospheric air, and due agitation; but this method is rarely adopted in practice. A high degree of heat decomposes the oxides again.