acid on sphalerite (zinc-blende), but further experiments have shown that when using an acid of sp. gr. 1.12, the action stops when about one-half of the acid is decomposed. In spite of this, the method is advantageous, and the residual acid liquid can be used for the evolution of hydrogen or carbonic anhydride. G. H. M.

Constitution of Acids. By W. A. DIXON (Phil. Mag., 21, 127130). Inorganic acids may be supposed to contain displaceable hydrogen, 1st, directly combined with the chlorous element, HCl, &c.; 2nd, combined as hydroxyl, CI(OH), N(OH), &c.; 3rd, as oxyhydroxyl, S(OOH)2, &c.; or partly in each form PH(OH)(OOH), P(OH),(OOH), S(OH) (OOH), PH (OOH). Acids of the first kind are usually feeble, and their salts have an alkaline reaction, except when they contain strongly chlorous elements. The group OOH confers high acidity, which is lost when the hydrogen is displaced by an alkaline metal.

Displacement of hydrogen in hydroxyl by an alkaline metal gives salts having an alkaline reaction. When an acid contains both OH and OOH, the hydrogen of the latter group will be first displaced; subsequent displacement of the hydroxylic hydrogen will give a salt having an alkaline reaction.

Silicic acid in solution (obtained by diffusion) may exist as Si2(OH)2(OOH)2, instead of Si(OH), and hence have an acid reaction. To meet certain cases, special assumptions must be made.

Ch. B.

Metals of Ancient Chaldea. By BERTHELOT (Compt. rend., 104, 265-271). The author has examined four tablets found in a stone receptacle under one of the corner stones of the palace of Sargon, which was built 706 B.C. One of the tablets consists of beaten gold, without any notable proportion of any foreign metal; the second consists of silver, almost equally pure, but more or less altered on the surface; the third has undergone considerable alteration, but the internal metallic portion is an alloy of copper and tin, which contains the two metals in the proportions 85.25: 10-04, and is therefore similar in composition to the majority of ancient bronzes; the fourth consists of pure crystallised magnesium carbonate, a somewhat rare mineral.

A fragment of a vase found during the excavations at Tello was found to consist of pure metallic antimony, and it is evident therefore that this metal was known to the ancients, although it is confounded with lead by Dioscorides, Pliny, and other writers. A small idol from the same excavations had undergone much superficial alteration, but the internal portion was found to consist of pure metallic copper. C. H. B.

Decomposition of Ammonium Chloride by Means of an Alloy of Zinc and Iron. By H. N. WARREN (Chem. News, 55, 3839).-A mixture of dry ammonium chloride and finely powdered zinc, alloyed with 12 per cent. of iron, was put into the right-hand limb of a tube, closed at one end, and bent like the letter N, some mercury or liquid amalgam of tin was put in the lower bend, the open end was

then sealed, and the mixture heated. At the termination of the reaction there was a great pressure of hydrogen, and a considerable quantity of ammonium amalgam in the tube. When the experiment was performed without pressure, or with lime instead of zinc, no ammonium amalgam was obtained. The zinc therefore gives rise to the separation of ammonium, which at first decomposes into ammonia and hydrogen, but as the pressure increases this decomposition no longer takes place, but instead the ammonium is made to amalgamate with the mercury. D. A. L.

Atomic Weights of Silver and Copper. By W. N. SHAW (Phil. Mag. [5], 23, 138-141).-The author determines the amount of copper deposited from a nearly saturated solution of copper sulphate by electric currents of different current density. He finds that the amount reaches a maximum at the highest limit (0·13 ampère) at which a coherent deposit is obtained. For current densities of 0.02— 0.00002, where d is the 0.0014 ampère the amount multiplied by 1 +

d

current density, equals that obtained with high current densities. From these values the author obtains for the atomic weight of copper the value 63.33 =

107.66 =

19 × 10

that of silver being taken as " 3

17 Си 10.

17 x 19
3

Ag hence =

Taking one-third the weight of

the hydrogen-atom as unit, the possible valencies of silver become 1, 17, and 19; of copper, 1, 2, 5, and 19. The only compound known to the author which supports this view is that formed between molten silver and oxygen, which has approximately the composition AgO. The author draws attention to the ratio between the atomic

weights of potassium and sodium.

17

10

H. K. T.

Decomposition of Sodium Sulpharsenate with Silver Nitrate. By K. PREIS and B. RAYMANN (Chem. Centr., 1887, 32).— Silver nitrate acts on sodium sulpharsenate in a different manner when added alone, or together with acid, to what it does when the action takes place in ammoniacal solution. In the latter case, silver arsenate and sulphide are almost entirely formed, together with a little arsenite, the quantity of which depends on the amount of free acid present. Since the reaction is instantaneous, the formation of the arsenite is perhaps due to reduction. The authors suggest the following explanation of its formation -Arsenic acid is a strong acid, whose thermal effect is equal to that of the strongest acids, whilst on the contrary arsenious oxide possesses feeble acidic properties, and even under certain conditions exhibits basic properties. Therefore if the reaction takes place in presence of ammonia, arsenic acid is formed, whilst in the presence of free acid the feebly basic arsenious trioxide is formed. This theory corresponds with the decomposition of silver nitrate by arseniuretted hydrogen; on passing this gas through

neutral silver nitrate solution, arsenious acid goes into solution, whilst if the silver solution is ammoniacal arsenic acid is formed.

G. H. M.

Tetracalcium Phosphate and the Phosphoric Acid of Basic Slag. By H. OTTO (Chem. Zeit., 11, 255-256).-The author's results confirm those of some other investigators; he finds the crystals separated from basic slag consist of tetracalcium phosphate, and moreover, that the phosphoric acid in the basic slag itself behaves with numerous solvents in the same way as tetracalcium phosphate; with alkaline ammonium citrate, however, this is not the case, owing to the interference of the silica and magnesia present in the slag.

D. A. L.

Properties of some Metals. By V. MEYER (Ber., 20, 497-500). -Magnesium melts at a temperature above the melting point of sodium bromide, melting at 700°, and slightly below that of anhydrous sodium hydroxide, melting at 800°. Experiments made with the object of determining the vapour-density of the metal have failed, inasmuch as it is impossible to volatilise magnesium at an incipient white heat in a neutral gas, such as hydrogen or carbonic oxide; combination of the elements ensues if nitrogen is employed, whilst carbonic anhydride is reduced to carbon by the metal at this temperature. Porcelain is strongly attacked by magnesium at an incipient. white heat.

Germanium does not attack porcelain at about 1350°, but is only volatile to a small degree when heated either in hydrogen or nitrogen at this temperature; its vapour-density, therefore, cannot yet be determined. Antimony is completely volatilised at 1300°, but not with sufficient rapidity to allow of its vapour-density being determined; further experiments will be made to determine this constant at a higher temperature. The author finds that mercury can be obtained free from metallic impurities by repeated distillation.

W. P. W.

Effect of Hydrochloric Acid on the Solubility of Chlorides. By R. ENGEL (Compt. rend., 104, 433-435).—The relation previously found to exist (Abstr., 1886, 505) between the solubility of chlorides and the proportion of hydrochloric acid in the liquid holds good generally, whether the chlorides belong to the same or different families, and whether they crystallise in the anhydrous or hydrated conditions. Experiments with magnesium and calcium chlorides show that in these cases the sum of the equivalents of the chlorides and hydrochloric acid in the liquid remains constant.

The author criticises Jeannel's conclusion (Abstr., 1886, 972), and considers that his results were due to the fact that the proportion of water added was relatively large compared with the quantities of metallic chloride and hydrochloric acid present.

The precipitation of the chlorides cannot be attributed to the combination of the hydrochloric acid with the water. The solubility of different chlorides is very different, and it would be necessary for the hydrochloric acid to combine with a different amount of water in each case. Both the water of crystallisation of the hydrated chlorides,

and the water combined with the hydrochloric acid play the part of solvents; the action is the same both for anhydrous chlorides and for hydrated chlorides calculated to the anhydrous condition.

Similar results have been obtained with nitric acid and nitrates, but with sulphuric acid and sulphates the phenomena are different.

C. H. B.

Pure Zinc, from the Bertha Zinc Company, Pulaski Co., Virginia. By G. B. BIRD (Amer. Chem. J., 8, 431).-By redistilling commercial spelter from a fire-clay retort with iron front, the amount of impurities is reduced with exception of the iron, which is scarcely diminished.

H. B.

Analysis of Shot. By H. HARDAWAY (Amer. Chem. J., 8, 432).— Several commercial brands of bird-shot were analysed, especially for arsenic; the amount found, namely, 04-30 part per 1,000, proved to be less than that usually stated (3 to 8 or 10 parts). H. B.

Preparation of Lead Carbonate. By W. KUBEL (Dingl. polyt. J., 262, 143).-Lead oxide is readily converted into the soluble hydroxide on treatment with a moderately concentrated solution of magnesium acetate. The solution thus obtained has an alkaline reaction, and yields lead carbonate when treated with carbonic anhydride. The white precipitate is collected, washed, and dried, and the solution of magnesium acetate concentrated and used for another operation. D. B.

Action of Lead Oxide on Soluble Chlorides. By G. ANDRÉ (Compt. rend., 104, 359-360).-If a concentrated solution of calcium chloride is boiled with an excess of lead oxide, care being taken to prevent evaporation, and the liquid is filtered, the filtrate deposits an abundant precipitate, which after washing with alcohol and drying has the composition CaCl2, CaO,2Pb0 + 4H2O. When boiled with water it becomes yellow. If the water is left out of account this compound may be regarded as crystallised calcium oxychloride in which 2 mols. of calcium oxide have been replaced by 2 mols. of lead oxide. When the mother-liquor, which is slightly alkaline, is poured into a large excess of cold water, the compound PbCl2,3PbO+ 3H2O is obtained as an amorphous white precipitate.

A solution of barium chloride when treated in a similar manner yields small slender needles which seem to be a compound of barium chloride with lead oxychloride, but they are decomposed immediately by water and yield lead oxychloride.

Strontium chloride yields small, distinct, brilliant crystals which when washed with alcohol and dried have the composition SnCl2,2PbO,5H2O. If the mother-liquor is poured into excess of water the compound PbCl2,3Pb0,3H,O is precipitated.

Lead oxide is dissolved by either a hot or a cold solution of magnesium chloride, but the filtered liquid deposits no precipitate, and only a very slight precipitate is obtained when the solution is poured into an excess of water. C. H. B.

Analysis of a Copper Slag of Bright Red Colour. By N. A. BERRY (Amer. Chem. J., 8, 429-430).-The slag is hard, breaks with conchoidal fracture, and contains small cavities and very minute metallic beads. Treated with hydrochloric and nitric acids, 30-23 per cent. dissolved (27.05 per cent. cuprous oxide), and there remained a yellow powder containing Cu,0 = 8·64; FeO 1169; SiO2 = Cu2O 49.24.

H. B.

Composition of Mirror Amalgam. By G. HARRISON (Amer. Chem. J., 8, 430-431).—The amalgam from a German mirror was found to contain tin 83.78 per cent., mercury 16.60 per cent.

H. B. Mercurous Hydroxide. By G. B. BIRD (Amer. Chem. J., 8, 426-427; compare Abstr., 1885, 124).-Solutions of mercurous nitrate and of caustic potash in alcohol of 70 per cent., were cooled to -42° and mixed; but little precipitate was formed and that slowly, its colour was a pale amber, changing with rise of temperature into the colours of mixtures of mercurous oxide, mercuric oxide, and finely divided mercury. Judging from the colours of the corresponding lead, copper, and silver compounds, it appears very probable that the pale amber-coloured precipitate obtained was mercurous hydroxide.

H. B.

Action of Mercuric Oxide on Dissolved Chlorides. By G. ANDRE (Compt. rend., 104, 431-433).-Klinger has described a compound, CaCl2,2HgO + 4H2O, and this the author has also obtained by the action of mercuric oxide on a solution of calcium chloride. The mother-liquor is not precipitated by cold water. If a solution of mercuric oxide in a boiling and concentrated solution of calcium chloride is poured into a large excess of cold water, a brickred precipitate is obtained which when dried at 100° has the composition HgCl2,3HgO. It is analogous to and possibly identical with the oxychloride obtained by Millon by mixing equal volumes of cold saturated solutions of mercuric chloride and potassium hydrogen carbonate.

A boiling solution of barium chloride, saturated with yellow mercuric oxide and allowed to cool, deposits needles of the compound BaCl2, HgO+6H2O, analogous to the barium oxychloride previously described. When treated with water, it becomes yellow. It loses nearly 5 mols. H2O at 100° and the rest at 150°. If heated in a tube, it gives off water, becoming yellow and finally white. Red mercuric oxide gives the same compound, which, when decomposed by water, yields the red oxide. If the mother-liquor of the oxychloride is poured into cold water, only a slight precipitate is formed.

Strontium chloride solution under similar conditions yields the compound SrCl2,HgO + 6H2O in long, slender needles, which lose 35 mols. H2O at 100° and the rest at 130°. It is analogous to strontium oxychloride but contains 3 additional mols. H2O. Only a slight precipitate is formed when the mother-liquor is poured into cold water.

A cold saturated solution of magnesium chloride dissolves yellow mercuric oxide when heated, but the liquid deposits no compound on