Laws of Chemical Equilibrium. By H. LE CHATELIER (Compt. rend., 106, 598-601).—The author points out that there is an essential difference between his views and those of Duhem.

C. H. B.

Explosion of Water Gas. By A. v. OETTINGEN and A. v. GERNET (Ann. Phys. Chem. [2], 33, 586-609).-The gas generated electrolytically was passed without previous drying into the eudiometer, and by means of a rotating mirror photographs of the course of the explosion were taken, small quantities of metallic salts being used to colour the flame. The results show that in accordance with Bunsen's view, the total combustion takes place in successive partial explosions. Each explosion causes a wave which travels along the tube, and the meeting of two such waves gives rise to secondary waves, as appears from the photographs. The rate of explosion measured was 2800 metres per second, which agrees approximately with the value obtained by Berthelot. H. C.

The Carbon-atom and Valency. By V. MEYER and E. RIECKE (Ber., 21, 946–956).—The theoretical conclusions of van't Hoff and Wislicenus have not only been supported but also extended by the experiments of Meyer and others on the isomeric benzildioximes; the result of which is, that to the hitherto generally accepted properties of the carbon-atom, the following two must be added:-(1) The four valencies of the carbon-atom can be diverted from the regular tetrahedric direction in which they are supposed to exist in marsh-gas, and compounds of the constitution Ca; (2) There are two ways in which two singly-bound carbon-atoms can be united, one which allows free rotation in various directions, one which does not.

On various chemical and physical grounds, the authors put forth the following hypothesis on the constitution of the carbon-atom :— The carbon-atom is surrounded by an ethereal shell which, in the case of an isolated atom, has a spherical form; the atom itself is the carrier of the specific affinities, the surface of the shell is the seat of the valencies; each affinity is determined by the existence of two opposite electrical poles, which are situated at the end-points of a straight line small in comparison with the diameter of the ethereal shell. Such a system of two electric poles is called a double- or dipole. The four valencies of a carbon-atom would be represented by four such di-poles, the middle points of which are situated on the surface of the ethereal shell, but freely movable within it. The dipoles themselves can rotate freely round their middle point. carbon-atom has a greater attraction for positive than for negative electricity, and the positive pole of a valency is slightly stronger than the negative pole.

The

This hypothesis would explain-Why the four valencies take up the position of a regular tetrahedron; why they can be diverted from this position; why the valencies of one and the same carbon-atom cannot combine together, whilst the valencies of different carbonatoms can do so; why there are two kinds of single-binding, (1)

one stable, (2) and the other allowing free rotation; and lastly, why free rotation ceases in cases of double- or treble-binding.

F. S. K.

On Valency. By H. E. ARMSTRONG (Phil. Mag. [5], 25, 21—30). The author extends Helmholtz's theory that the valency of an atom corresponds with the number of charges of electricity with which it is associated (Trans., 1881, 277) by supposing that "molecular" combinations are due to the same cause. As instances of substances which may be looked on as molecular compounds, the author quotes the tetralkyl-compounds of ammonium. The stability of the iodides of these bases in the presence of potassium hydroxide is such as is shown by no other iodide except those of carbon; they fully resemble in this respect methyl iodide and iodobenzene. Moreover, although the heat of neutralisation of tetramethylammonium hydroxide is equal to that of potash whilst that of methyl alcohol is small, in the latter case the combination is incomplete and in many cases the salt produced is insoluble and does not interact with water. In aniline, the basic properties, so far from being entirely dependent on the nitrogenatom, are lessened and ultimately almost annulled by the introduction of halogens into the phenyl radicle. Some hydrazines are monobasir, others bibasic, although they all contain two (triad) nitrogen-atoms: hence, the pentad nature of the nitrogen in the salts of these substances is doubtful, and they are better considered as so-called molecular compounds. To explain these molecular combinations, the charge of electricity is supposed to penetrate the atom and to have varying degrees of freedom of motion. In some combinations, such as H2, the opposite charges approach one another, and there is no residual affinity; whilst in others, such as HCl, the charges of some of the atoms do not move towards those of their neighbours, their influence is not entirely neutralised and a residual affinity obtains. In tetramethylammonium iodide, the nitrogen of the molecule NMe, has such a residual affinity which is neutralised by a similar residual affinity in the iodine of the methyl iodide. Ammonium haloïd compounds, in that they can further combine with halogens, show residual affinity, but it is not probable that the halogen of the ammonium salt serves as the bond of union. On the other hand, in the double metallic salts the halogen does play this part, and from their analogy to ammonium haloïd compounds the halogen in the latter should serve as the bond of union. In view of this conflict of evidence, the theory of the ammonium radicle is questioned, and the perhalogen-compounds are regarded as formed by the union of the halogen with the nitrogen-atom -a view supported by the fact that ammonium sulphate combines with halogens. That the same result is obtained in whatever order the radicles are inserted in a tetralkylammonium-compound may be due to isomeric change at the moment of formation, and does not necessarily prove the atomic constitution of these compounds. In the case of phosphorus, the existence of the compound PF, does not prove its quinquavalence in face of the compounds HF, and HFFR, which prove that fluorine can combine with itself. Instead of the constitution Cl2P OCI for phosphorus oxychloride, the oxygen may be considered

to be united to the phosphorus-atom in PCl, by residual affinity. For bismuth, the compound BiPh, Br2 only proves that it belongs to the nitrogen-phosphorus-group so long as its constitution is unknown. In the compounds PbEt, and TeCl, the lead and tellurium atoms may be considered to have charges showing little mobility, hence each charge binds two monad atoms. The elements of the carbon-silicongroup are probably tetrads, whilst iron in the ferric compounds is a triad. Chlorides in which the chlorine-atoms form a closed chain are also conceivable. From the above it follows that, except in the paraffins, benzenes, and their haloïd and alcoholic derivatives, the constitution of compounds is by no means settled, and that a more liberal interpretation of facts than heretofore must be made. H. K. T.

Tetravalency of Oxygen. By J. F. HEYES (Phil. Mag. [5], 25, p. 221-237). The author considers that molecular compounds of haloïd salts can be better explained by supposing the haloïds to be trivalent than by grouping the atoms round a central atom, which must, therefore, have a very high valency. This, with the occasional tetravalency of oxygen, gives formula for the oxychlorine acids more in accord with their properties than those in use. Thus potassic hypochlorite becomes K-C1-0—0—Cl-K, whilst the substances KČIO3, KCIO, may be looked upon as K-Cl- united to groups of oxygen-atoms (O3)" or (0.)", forming perhaps a closed chain.

The tetravalence of oxygen is supported by the peroxides, BaO2, MnO2, whose properties, distinct from those of SnO2,SiO2, point to a composition X"0-0, and by the oxides of silver, mercury, and copper. Sodium dioxide, since it is not decomposed by heat, is considered to have the structure Na-0-0-Na, whilst in K2O, there is an (0) nucleus related probably to that in KCIO, and Cl2O. The interaction of water with organic substances such as aldehyde, acetal and acetic acid, favours the view of the occasional tetravalency of oxygen, as do also the gaseous acetic acid molecule (C,H,O,) of Ramsay and Young, and Friedel's substance, (CH3)2O, HCl. The double metallic oxides, basic salts, and water of crystallisation (—OH2·OH2·OH2—, &c.) also favour this view. In substances where there are two valencies the author proposes the term validity to indicate the less powerful affinity: thus chlorine is monovalent and trivalid; carbon is tetravalent, except in CO and CNO, where it is bivalid. J. J. Thomson has pointed out that according to the vortex theory a dyad might unite not only with two atoms but also with four atoms of a monad, so that water consists of three primaries H2-0-0. The author's formula, H20=0, agrees with this theory and also with Brodie's H,00. As an instance of a radicle at once mono- or tri-valent, the author quotes (CH)", which if regarded as an atom is analogous to gold or thallium. H. K. T.

+

Investigation of the Second van't Hoff Hypothesis. By K. AUWERS and V. MEYER (Ber., 21, 784-817).—This vol., p. 597.

Simple Procedure for the Determination of Molecular Weights by Raoult's Method. By A. F. HOLLEMANN (Ber., 21, 860-862). The simplifications introduced proceed on the same lines as those already described by V. Meyer and by K. Auwers (this vol., pp. 407-409).

Ancient Process for making Gems and Glasses Phosphorescent. By BERTHELOT (Compt. rend., 106, 443-446).-An extract from ancient Greek MSS. containing recipes for making gems, glasses, &c., phosphorescent, or imparting to them more brilliant colours, by coating them with thin layers of various-coloured or phosphorescent substances, the latter being usually of the nature of readily oxidisable organic compounds. C. H. B.

Inorganic Chemistry.

Rate at which Bleaching Powder loses its available Chlorine when kept at Different Temperatures. By J. PATTINSON (J. Soc. Chem. Ind., 7, 188--191).-It is shown that although the available chlorine of bleaching powder disappears in proportion to the length of time it is kept and to the temperature to which it is subjected, yet, within a comparatively small amount, the whole of the chlorine it originally contained is still retained in some form or other in combination with the calcium of the compound. This, at any rate, is true of bleaching powder when kept in casks or bottles for about 12 months at temperatures varying from 15.5° to 26.5°. D. B.

Preparation of Tetrathionates from "Wackenroder's " Solution. By T. CURTIUS and F. HENKEL (J. pr. Chem. [2], 37, 137 -149; compare Debus, Trans., 1888, 278).-When "Wackenroder's " solution is shaken for a considerable time with excess of barium carbonate and then filtered, the solution is quite clear, neutral to litmus, and contains sulphur and barium present in the ratio of four atoms of the former to one of the latter, thus corresponding with barium tetrathionate. If, however, the solution is imperfectly neutralised it will contain acid tetrathionates, and consequently sulphur will be present in a quantity larger than that represented by the ratio S. Ba. The existence of these acid tetrathionates, unknown until now, has led many investigators to wrongly assert that polythionic acids, containing more sulphur than tetrathionic acid, are present in Wackenroder's solution. Pure barium tetrathionate may be obtained from Wackenroder's solution, which has been completely neutralised by barium carbonate, by precipitating with alcohol and then alternately dissolving in water and precipitating with absolute alcohol. As thus prepared, the salt is identical with that formed by the action of iodine on the thiosulphate, and has the formula BaS,O + 2H2O.

If Wackenroder's solution is neutralised by means of the carbonates of zinc and manganese, and then an equal volume of the solution added, acid salts of these metals are obtained. These acid salts are much more stable than the normal ones, and dissolve readily in cold absolute alcohol, from which they can be recrystallised over sulphuric acid in a vacuum. The zinc salt, (S,O,H),Zn, crystallises from alcohol in groups of small needles, is somewhat hygroscopic, and is exceedingly soluble in cold water and in cold alcohol. It decomposes at a little above 100°, and addition of potash to the aqueous solution causes an immediate separation of sulphur. The manganese salt, (S,O,H), Mn, crystallises in rose-coloured plates, and in all its other properties much resembles the zinc salt. The authors have observed the peculiar appearance and properties of the sulphur present in Wackenroder's solution (Debus, loc. cit.). G. T. M.

Hydrofluorides of Potassium Fluoride. By H. MOISSAN (Compt. rend., 106, 547-549).-When dry, powdered potassium fluoride is gradually added to liquid anhydrous hydrofluoric acid it dissolves rapidly with development of heat, and if the solution is cooled to -23° it yields highly hygroscopic crystals of the hydrofluoride of potassium fluoride, KF,3HF. This compound gives off hydrogen fluoride in moist air, and dissolves rapidly in water with liberation of acid and great reduction of temperature. When heated, it loses hydrogen fluoride and yields a residue of potassium fluoride. At 100°, the fused salt has no action on crystallised silicon, but if rapidly heated to a higher temperature the mixture becomes incandescent with violent evolution of silicon fluoride. The fused salt acts energetically on silica and decomposes carbonates. It is decomposed by sulphuric acid in the cold with very rapid evolution of hydrogen fluoride, and acts energetically on potash and ammonia.

If the potassium fluoride and hydrogen fluoride are mixed in suitable proportions, the compound KF,2HF is obtained; this is liquid at 105°, and on cooling forms crystals similar to those of the preceding compound. It may be kept liquid between 65° and 105°, and under these conditions acts readily on many minerals and organic sub

stances.

These compounds are analogous to Berthelot's hydrochlorides of chorides and Troost's compounds of ammonia with ammonium salts. The trihydrofluoride is, however, somewhat stable, and decomposes very slowly in dry air or even in a vacuum. C. H. B.

Preparation of Cakes of Ammonium Chloride and Ammonium Carbonate. By W. HEMPEL (Ber., 21, 897).-When heated at 50-100° and subjected to hydraulic pressure, powdered ammonium chloride and ammonium carbonate are respectively converted into hard cakes. W. P. W.

Soluble Phosphates in Superphosphates. By H. OTTO (Chem. Centr., 1887, 1563-1564; from Zeit. Chem. Ind., 2, 207-210).The solubility of the pure calcium salt, CaH (PO4)2, in water at 15° is about 1 in 25. Solutions of the phosphate do not undergo decomposi