The

effected by an attraction acting along and represented by the straight line joining them. By such a representation, however, a double union would be in no way distinguishable from a single one, both deing denoted by a straight line joining the carbon-atoms. author considers this to indicate a weak point in the Van't Hoff - Le Bel hypothesis, and one requiring explanation. He also concludes that a polyvalent atom cannot be represented as a point in space, but that various portions of such an atom having different affinities must be distinguished. H. C.

Inorganic Chemistry.

Preparation of Hydrogen Iodide. By L. MEYER (Ber., 20, 3381-3383).—100 parts of iodine contained in a retort are moistened with about 10 parts of water. The retort is then fitted with a funnel closed with a glass rod, containing 5 parts of amorphous phosphorus mixed with 10 parts of water. One drop of water containing pbosphorus is let into the retort; more phosphorus is slowly added, after which large amounts may be added. The mixing is completed in a quarter of an hour. If more than a drop is added at first, the action cannot be controlled, and will generally result in a violent explosion. No heating is necessary. The iodine carried over by the hydrogen iodide is nearly all deposited in the neck of the retort, which is inclined upwards.

By using 100 grams of iodine, 5 grams of phosphorus, and 25 c.c. of water, 95 grams of hydrogen iodide (of which 37.5 grams were obtained by distillation) were obtained, instead of 1008 grams. With 20 grams of water 98.1 grams were obtained (744 grams as gas and 237 grams by distillation).

N. H. M.

Products and Rate of Decomposition of the Salts of the Halogen Oxy-acids by Heat. By A. POTILITZIN (Chem. Centr., 1887, 1218, from J. Russ. Chem. Soc., 1887, 339-357).-Between 330° and 400°, barium chlorate decomposes entirely according to the equation 2Ba (ClO3)2 = BaCl2 + Ba(CIO)2 + 40. Between 400° and 470°, barium chloride and oxygen are the only products. The rate of the decomposition rises with the temperature up to 400°, whilst if the temperature remains constant, the decomposition rises at first, and then gradually falls again, although it is not complete. According to the author, barium chlorate decomposes to some extent before it melts. Barium perchlorate crystallises with 3 mols. H2O, whereas Marignac gives the formula Ba(CIO,)2 + 4H2O; also, contrary to Marignac's statement, it is not hygroscopic. It loses 2 mols. H2O by remaining over sulphuric acid, and the third molecule is expelled at 100°.

Barium bromate becomes anhydrous at 170°, the crystals turning

slightly yellow, without, however, losing their crystalline form or lustre. Decomposition begins at 260°, and at 300° bromine is evolved. The decomposition becomes complete without the salt melting. The author draws the conclusion that the bromate is changed into two isomeric salts, namely BrO,OM (the original salt), and MBO, and that these, therefore, decompose at different rates. He could not find any perbromate in the product of decomposition.

J. W. L.

Mutual Displacement of the Halogens in their Compounds with Oxygen. By A. POTILITZIN (Chem. Centr., 1887, 1218-1219; from J. Russ. Chem. Soc., 1887, 358-364).—The author believes that these reactions are much more complex than is usually supposed. By the action of chlorine on a solution of sodium bromate in the dark, he obtained a mixture of sodium chloride and bromide, and bromic and chloric acids together with free bromine. He finds that chlorine acts in the same way on potassium bromate and on barium bromate ; experiments were also made on the dry salts, when the same products seemed to be formed. Bromine was also found to act on the chlorates in the same way, although much less energetically. These reactions take place much more readily in a sealed tube. Iodine and the iodates were also included in the series of experiments with similar results. J. W. L.

Preparation of Hydrogen Sulphide Free from Arsenic. By C. WINKLER (Zeit. anal. Chem., 27, 26-27).-Barium sulphate (powdered barytes) is mixed with 25 per cent. of ground coal and 20 per cent. of common salt. The damped mixture is rammed into a clay crucible, which after drying and closing with a luted cover is heated for several hours at an incipient white heat. The product is in hard compact masses which dissolve completely in dilute hydrochloric acid with a steady evolution of hydrogen sulphide.

M. J. S.

Selenites. By BOULZOUREANO (Bull. Soc. Chim., 48, 209-210).— Solutions of metallic salts were mixed with sodium selenite and the resulting precipitates were heated in sealed tubes at 200° with very dilute selenious acid. Ferric selenite forms small, golden-yellow prisms which separate in radiating groups.

Another method consists in treating a metallic carbonate with dilute selenious acid, mixing the solution with its own volume of water, and heating in sealed tubes at 200°. This was employed for the preparation of the cobalt, nickel, manganese, and cadmium salts.

Cobalt selenite forms transparent, violet prisms, nickel selenite forms large, short, green prisms arranged in radiating groups. If the original solution is mixed with sodium selenite, yellow crystals are obtained. Manganese selenite crystallises in short, pale-red prisms or in slender, brown needles if heated beyond 230°. The cadmium salt forms long, colourless, transparent prisms, or much shorter yellow crystals if heated at about 200°. Zinc seems to form a selenite crystallising in prisms.

The liquid obtained by the second method was allowed to evaporate spontaneously at the ordinary temperature or in a vacuum. Crystals

are obtained in the case of cobalt and manganese. The latter yields a pink-coloured crust and the former deep violet crystals.

Cupric carbonate, when treated with a warm solution of selenious acid, yields a blue precipitate of the normal selenite, and if this is heated to boiling, it is converted into green, microscopic prisms. The solution is bluish-green, and on cooling deposits large, green crystals. If the normal selenite is heated in sealed tubes with the carbonate and water, it yields well-defined, greenish-yellow, prismatic crystals. Sodium selenite and zinc sulphate in sealed tubes yield white, transparent crystals. C. H. B.

Preparation of Hydrogen Arsenide. By A. CAVAZZI (Chem. Centr., 1887, 1097, from Rend. Acc. Bologna, 1886-87, 85–86).— The action of zinc on an acid solution of arsenious acid produces a gas containing 70 per cent. by volume of hydrogen arsenide. Sodium amalgam containing not more than 4 grams of sodium in 50 c.c. of mercury, by its action on a concentrated solution of arsenious acid, produces a gas containing 86 per cent. by volume of hydrogen arsenide. A gas containing a large quantity of arsenic may be prepared by the action of aluminium on a somewhat dilute alkaline solution of potassium arsenite, whilst a solution of arsenic disulphide in potash, when subjected to the action of aluminium, evolves a gas quite free from arsenic. J. W. L.

Action of Hydrogen Arsenide on Arsenious Acid. By D. TIVOLI (Chem. Centr., 1887, 1097, from Rend. Acc. Bologna, 1886-87, 98). The reactions of hydrogen arsenide on arsenious acid dissolved in hydrochloric and sulphuric acid respectively are expressed by the equations:-2ASH, + 2AsCl, = 6HCl + As, and 3(AsO2) SÖ1 + SÕ ̧ H2SO,+6ASH, 3A,+ 4H2SO, + 6H2O. The precipitation of the arsenic does not take place in neutral solutions, whereas in the hydrochloric acid solution it is complete, and in the sulphuric acid solution almost so. J. W. L.

=

Lowest Compounds of Silver. By O. v. D. PFORDTEN (Ber., 20, 3375-3381; compare Abstr., 1887, 699).-According to Stas, silver is oxidised in the cold in presence of water containing dissolved air and acidified with hydrochloric, sulphuric, or acetic acid, &c. Experiments made by the author show that when a solution containing potassium permanganate and sulphuric acid is boiled in a current of carbonic anhydride it is not capable of dissolving finely divided silver; on admitting air to the solution, silver dissolves and the permanganate becomes decolorised. This accounts for the result previously obtained (loc. cit.) when silver was boiled with dilute sulphuric acid and treated with a drop of permanganate solution; the solution being free from air, the permanganate did not decolorise. In presence of air, the reaction is very slow, and in the titration of the argentous oxide with permanganate, none of the silver formed is dissolved (compare Friedheim, Abstr., 1887, 1079).

With regard to Friedheim's supposition (loc. cit.) that the argentous oxide is a mixture of silver and argentic oxide or organic matter, it is

mentioned that the preparation previously described contained no carbon and dissolved in nitric acid without leaving a residue; when shaken for 12 hours with mercury, it underwent no change in properties or appearance. It is therefore maintained that the substance cannot be metallic silver. The examination of the substance will be continued. N. H. M.

Behaviour of Basic Slag with Water charged with Carbonic Anhydride. By M. A. v. REIS (Chem. Zeit., 11, 933-934; 981982). In these experiments, which are a continuation of those previously made by him (Abstr., 1886, 663), the author has included nine samples of slag from different sources; two samples of precipitated slag; one sample each of di-, tri-, and tetra-calcium phosphate (the last synthetically prepared), also bones, bone-ash, and phosphorite. Ten grams of the finely powdered phosphate supported on a platinum cone covered with asbestos in a funnel, were treated with water saturated with carbonic anhydride percolating at the rate of 1 litre in three hours. One slag was treated with 50 litres of carbonic anhydride solution; each litre of the first 10 litres of percolate was examined separately, but subsequently examinations were made at definite intervals. Silica and phosphoric acid pass through in nearly constant quantities in proportions approximating to their relative amounts in the original slag up to the 25th litre, after which their solubility rapidly decreases; lime, however, continues to dissolve even in the 50th litre. Although the various changes take place simultaneously at first, the free lime is most readily attacked, then the calcium silicate and phosphate, and finally the other compounds of calcium (probably ferrite or manganate). The exhausted slag amounted to 30:07 of the original; its composition (I), and that of the original slag (II), is as follows:

The slight solubility of the magnesia is noteworthy. The other phosphates and slags were treated each with 10 litres of carbonic anhydride water, and the percolates were examined in two lots of 5 litres each. Full numerical details are given, from which the following table is taken, showing the relative solubility of the phosphoric acid, silica, and lime in the various phosphates examined; the figures are percentages of the amount of each constituent in the original

The slags varied considerably amongst themselves owing to their different origin and constitution, but along with the tetracalcium phosphate they proved more soluble than the other forms of phosphate; the comparatively high solubility of bone phosphoric acid is due to the organic matter present. It is interesting to note that in experiments with the tetracalcium phosphate after the removal of any excess of lime in the first 5 litres or so, the phosphoric acid and lime wash out in the molecule ratio IP2O: 4CaO, tending to show that this really is a compound of that composition, for had it been a mixture of lime and triphosphate, the lime would have continued to wash out irregularly. În a similar manner, the existence of tetracalcium phosphate in basic slag is illustrated.

The author concludes that he has now set aside v. Maltzahn's adverse criticism of his previous work, and upset any views as to the comparative insolubility of basic slag phosphates. D. A. L.

Tetrabasic Calcium Phosphate and the Basicity of the Silicate in Basic Cinder. By G. HILGENSTOCK (Chem. Centr., 1887, 1097-1098, from Stahl u. Eisen, 7, 557-560).—The author has succeeded in preparing tetrabasic calcium phosphate by fusing together calcium phosphate or phosphoric acid with lime, using fluorspar as a flux. The author points out further that since calcium triphosphate is reduced by metallic iron when fused, it can only be the tetrabasic phosphate which is contained in the basic slag, and that the difference in crystalline form of this phosphate may be accounted for in somewhat the same way as in the formation of the various modifications of antimonious oxide, valentinite and senarmontite. Observations seem to show that as the flux cools, the rhombic plates are first formed, next the hexagonal needles, and finally the monosymmetrical crystals. In conclusion, the author endeavours to prove that the silica can only be present in the form CaSiO3.

J. W. L.

Behaviour with the usual Solvents of the Soluble Phosphoric Acid in Superphosphates, which have remained some time in Bulk. By A. BEYER (Chem. Centr., 1887, 1115, from Rep. anal. Chem., 7, 327-330).—In making a series of determinations to compare Petermann's and Wagner's methods for the estimation of the phosphoric acid soluble in ammonium citrate, the author found that not only was. the percentage of phosphoric acid soluble in water reduced by long standing in heaps, but also that soluble in ammonium citrate solution as estimated by Wagner's method. The alteration in the percentage