General and Physical Chemistry. Validity of Maxwell’s Equations. By P. S. WEDELL-WEDELLSBORU (Zeit. physikccl. Chern., 1897, 24, 367-370).-Known facts are, in the author’s opinion, not in accord with Maxwell’s equations, the validity of which is contested, for according to these equations, there must be, at the starting of a current, two induction effects: (1) that experi- mentally obtained; (2) that due t o the increase of the electrostatic field. The paper also contains the author’s reply to some of Gold- hammer’s criticisms of a previous paper on this subject (Zeit. physikal. Chenz., 23, 686). By WILLIAM SKEY (Chem. News, 1897, 76, 209).-In the following couples, the electrolytes were separated by gelatin. With gold in potassium cyanide, and platinum in an acid, hydrogen is evolved from the platinum, but if an alkali is substituted for the acid, this is not the case.With gold or silver in cyanide, and gold or silver in copper sulphate, copper is deposited. With platinum or gold in tannic acid, and potash and platinum or gold in acid, a strong current is developed, and hydrogen is evolved until all the tannic acid is oxidised. Platinum in potash is positive to platinum in acid or ferrous sulphate, and throws down gold or platinum or silver from solution. Platinum in concentrated salt solution rendered alkaline with potash, and platinum inacid also throws down gold, from its chloride, on platinum and produces a strong current. An insulated voltaic cell connected with insulated silver plates 6 inches square and an inch apart in the air develops a current too feeble for detection by the galvanometer, but deposits gold, from its chloride, on platinum, Thermodynamics of “ Swelling” (“ Quellung”), with Special Reference to Starch and the Determination of its Molecular Weight.By HERMANN RODEWALD (Zeit. physikccl. Chern., 1897, 24, 193-218. Compare Abstr., 1895, i, 165).--The author applies the term ‘‘ swelling ” to the absorption of water, or other liquid, by a solid substance which does not exhibit any pores, visible or microscopic. An expression is deduced thermodynamically for the heat liberated in terms of the change of specific volume, namely, (1) y = F[dp/dt(s’ - s - )y], and the author then records the experimental work for the case of starch. Air-dried starch mas found to contain about 16.33 per cent.of moisture, and the heat produced by soaking in water was observed for starches of different original water content, an ice-calorimeter being employed ; a curve and table of the results are added. The contraction in volume mas also obtained as a function of the percentage of water, the curve being very similar to the above heat curve. The values of dp/dt and y in the equation (1) are deduced from the values of the heat and volume change for the dry and moist starch, and the expression y = 373(s’ - s ) l 75 - 0.00507 is thus obtained. A relation between the volume change, hence the heat of swelling, and the water content is next calculated, from which the equation log(s’ - s) = 0% - 2 - 0 . 0 4 2 3 ~ is ob- L. M. J. Laboratory Notes. D. A. L.VOL. LXXIV. ii. 562 ABSTRACTS OF CHEMICAL PAPERS. tained, and hence y = 0 when w = 31-63, By means of an expression used by Kirchhoff for the vapour pressure of sulphuric acid solutions, that of the starch is calculated, the value thus obtained being 4,5594 for 31*63 per cent. of water; from this calculated value of the vapour pressure, the author deduces the number 4370 for the molecular weight of starch corresponding with the formula C162H2700135. The expansion coefficient was found to be a linear function of the water, and the force of attraction between dry starch and moisture is calculated as 2073 kilos. per sq. cm. By c'. H. BENEDICT (J. Physical Chem., 1897, 1, 397--402).--The author finds that there is an enormous increase in the volatility of solid naphthalene when ether is present.Distillation with a current of ether vapour would give not less than four times as much naphthalene as if the process were carried on with an air current or under diminished external pressure. Some rough measure- ments with camphor in different solvents gave similar results. L. M. J. Distillation with Vapour. H. C. Solubility of Solids in Vapours, By J. M. TALMADGE (J. Physiccd eltern., 1897, 1, 547--554).--The author distilled saturated solutions of camphor and naphthalene in methylic alcohol, ethylic alcohol, acetone, and ether under different pressures, an excess of the solid being present in all cases. I n the case of naphthalene, he was unable to confirm Benedict's result (see preceding abstract), and although the experiments show that the concentration of naphthalene vapour in equilibrium with solid naphthalene is not independent of the other components in the system, they do not show whether the vapour pressure is increased or decreased by the presence of a solvent.I n the case of camphor, the vapoixr pressure of the solute also varies with the solvent. The values with ether and acetone are more than double the real vapour pressures. With methylic alcohol, the values are a little above the normal, whilst with ethylic alcohol, the calculated vapour pressures are only about one-half of those obtained directly. Temperature of Maximum Density of Barium Chloride Solutions. By Lours C. DE COPPET (Compt. rend., 1897,125, 533)- The following results were obtained. H. C. Weight of Reduction of the Molecular reduction BaC1, in 1000 grains Temperature of temperature of of temperature of of water.maximum density. maximum density. maximum density. 3,982' - - 0 6.73 3.207 0.775O 23.94 10.42 2.785 1-197 23.88 20.83 1.572 2.409 24.04 41.72 - 0.843 4.825 24.04 The reduction of the temperature of maximum density is sensibly pro- portional to the weight of anhydrous barium chloride dissolved in 1000 grams of water. By A. AICNAN and E. DUGAS ( C o q ~ t . Tend., 1897, 125, 498-500).-Mixtures of acetic acid and benzene in various proportions become homogeneous at different temperatures. C. H. B. Solubilities of Liquids.GENERAL AND PHYSICAL Acetic acid ... ... 30 C.C. 40 C.C. Benzene ............ 70 C.C. 60 C.C. Temperature of } homogeneity 75" CHEMISTRY.63 50 C.C. 60 C.C. 70 C.C. 50 C.C. 40 C.C. 30 C.C. 50" 3 0" 3.5O and in these cases it is difficult t o decide which is the solvent and which the dissolved substance, as in: the case of aniline, phenol and qvater described by Alexkeff (Abstr., 1886, 847), and his method of observation does not admit of the solution of the problem. If V, and Vb are the respective volumes of the two liquids A and 13 introduced into the sealed glass tube, a and p their coefficients of reciprocal solubility, and Vl and V2 the respective volumes of the two layers of liquid, A saturated with B and B saturated with A, and a similar relation holds good if weights are taken instead of volumes. When the temperature varies, if a tends towards the value V,/Yb, the numerator of the first member Vl should tend towards zero ; but if /3 tends towards the value Val Vb, then V2 should tend towards zero.It becomes necessary, therefore, t o observe towards which extremity of the tube the surface of separation of the liquids disappears on heating. I n the cases quoted, with 30, 40, or 50 per cent. of acetic acid, the surface of separab'ion tends towards the bottom of the tube, and hence a t 75" and 50" saturated solutions of acetic acid in benzene occur; with 60 and 70 per cent. of acetic acid, the surface of separation tends to disappear towards the upper part of the tube, and hence at 30° and 2-5" saturated solutions of benzene in acetic acid occur. The authors criticise Alexeeff's method of drawing his curves, which they regard as based on an erroneous assumption.C. H. B. Kinetic Theory of Solutions. By ARTHUR A. NOYES (Zeit. physikul. Chem., 1897, 24, 366).-In his paper on osmotic pressure (Abstr., 1897, ii, 395), the author erroneously ascribed an inaccurate expression to Nernst. L. M. J. Osmotic Pressure and Variance. By JOSEPH E. TREVOR (J. Pl~ysiccd C'lem., 1897, 1, 349-365).-The variance is always the total number of variables which a system exhibits, diminished by the number r of its phases. The variables are n potentials, being at least one for each' independently variable component, one for the temperature, and one for a t least one pressure, with x for the added pressures introduced by the appearance of x osmotic walls, and y for each of the y separa- tions of a component by such a wall. The total number of variables is, therefore, n + 2 + n: + y, and the variance is v = n + 2 + x + y - r.This is the generalised phase rule as applicable to all systems contain- ing osmotic pressures. For the limiting case in which all osmotic walls are absent, we have both x = 0 and y = 0, and consequently the Gibbs' variance of v = 1% + 2 - Y. H. C. 5-264 ABSTRACTS OF CHEMICAL PAPERS. Variance of Osmotic Systems. By JOSEPH E. TREVOR (J. yhy- siccd Chem., 1897, 1, 537-541).-The author shows how it is possible to renlise for two component monovariant systems the anticipated set o€ five curves of osmotic pressures, and their intersection at a five-fold multiple point (compare preceding abstract). The Phase Rule and the Physical Properties of Chemical Compounds. By F.WALD (Zeit. physikal. Chem., 1897, 24, 3 15-334).-The author discusses the well-known phase-law of Gibbs, namely, v = n + 2 - Y, where r is the number of phases, v the number of independent variations, and 9% that of the independent components, two physical conditions only changing. The case chiefly considered is that for an equal number of phases and components, and the author regards i t as proved that a number of physical relations ‘‘ may be deduced between the substances entering into a reaction, especially Gay Lussac’s By WILDER D. BANCROFT (J. Pl~ysiccd Clem., 1897, 1, 337-343).-1n a system composed of two salts and water there will be in equilibrium at the quintuple points three solid phases, solution, a i d vapour. The various quintuple points can be classified under three heads.I. Two of the solid phases can be made from the third with addition or subtraction of water. 11. One of the solid phases can be transformed into one of the others by addition or subtraction of water. 111. No one of the solid phases can be converted into either of the others by addition or subtraction of water. When one of the solid phases can change into the other two with addition or subtraction of water, the inversion point is a minimum temperature for that phase if the water be added to complete the re- action, and a maximum if the water be subtracted. If one of the solid phases can be converted into one of the others by addition of water, the inversion point is a maximum or a minimum temperature for one of those phases, and is neither a maximum nor a minimum for the third solid phase.When no one of the solid phases can be con- verted into either of the others by addition or subtraction of water, no prediction can be made. There cannot be in equilibrium three solid phases such that one can be made from the other two without addition or subtraction of water. H. C. Solids and Vapours. By WILDER D. BANCROFT (J. Physical Chem., 1897, 1, 344--348).-Whilst many salts in efflorescing at con- stant temperature form all intermediate hydrates, this is not always the case. For example, Na,S04+10H20 changes normally to the anhydrous salt without formation of Na2S04+7H20. If we start with hydrated sodium sulphate, solution, and vapour, and raise the temperature t o about 3 3 O , the anhydrous salt will be formed.On decreasing the external pressure, the solution will disappear, leaving the stable monovariant system, hydrated and anhydrous sodium sul- phate and vapour. Decreasing the pressure yet more, Na,S04 + 1 OH,O must effloresce with formation of the anhydrous salt. Were it to form H. C. gas law of rational volume ratios.” 1,. M. J. Quintuple Points.GENERAL AND PHYSICAL CHEMISTRY. G 5 the heptahydrated salt, there would be present a non-variant system. This would also be formed from the monovariant system a t any other temperature a t which the latter could exist, and we should thus have the phenomenon of a non-variant system existing a t a series of tem- peratures and pressures, which is impossible according to the phase rule. From this it follows that a solid phase containing two com- ponents effloresces with formation of the solid phase which can co-exist at the next higher quadruple point.Two solid phases containing three components effloresce with formation of the solid phase which can co-exist a t the next higher quintuple point. From a study of the efflorescence products, one can draw coiiclusions as t o the phases exist- iag a t the quintuple points. A Triangular Diagram [to represent Composition-Tempera- ture Changes]. By WILDER D. BANCROFT (J. Physicul Chenz., 1897, 1, 403--410).-A diagram consisting of an equilateral triangle with lines ruled parallel to each side, instead of perpendicular t o them, was proposed by Roozeboom for the representation of the changes in com- position of a given phase with the temperature when there are three components.The author points out some geometrical relations con- nected with the use of this diagram. Two Liquid Phases. By WILDER D. BANCROFT (J. Plqsicul Chem., 1897, 1, 414--425).-The author considers the general case of quintuple points with two solid phases, two liquid phases and vapour, formed by adding a component C to two components A and B such that there can be formed the quadruple point, solid A, two solutions, and vapour. The freezing point rises. The solid phases a t the quintuple point are A and C or else no non-variant system with two liquid phases is possible. The freezing point falls. There is one quintuple point with A and C as solid phases, or two with A and B, B and C as solid phases, or one with A and B as solid phases.3. The component C increases the miscibility of A and B ; the freezing point falls. There is one quintuple point with A and C as solid phases, or one with A and B as solid phases, or there is formed the divsriant system, solid A, solution, and vapour. 4. If the component C dissolves in A with precipitation of B and there are two quintuple points, the one with B and C as solid phases mill exist a t a higher temperature than the one with S and B as solid phases. 5. If the component C increases the miscibility of A and B and there are two quintuple points, the one with B and C as solid phases exists a t a lower temperature than the one with A and B as solid phases. H. C, Solubility and Freezing Point. By DOUGLAS MCINTOSH (J. Physicc~l Chem., 1897, 1, 474-492).-When we have two non-miscible substances A and C and a third substance B with which the other two are miscible we can distinguish two cases.H. C. H. C. The general results are. 1. The component C dissolves in B with precipitation of A. 2. The component C dissolves in A with precipitation of B.66 ABSTRACTS OF CHEMICAL PAPERS. I n the firsf case, the component A can exist as solid phaseunder the conditions of the experiment. Under these circumstances, addition of C to the liquid phase containing A and JT) will raise the temperature at which A can exist as solid phase. I n other words, addition of a sub- stance to a binary solution in equilibrium with a solid phase raises the freezing point if the substance added be non-miscible with the corn- ponent appearing as solid phase.This is shown to be the case for an alcohol-benzene solution to which water is added. In thesecond case, the component R can exist as solid phase under the conditions of the experiment. Under these circumstances, addi- tion of B to the liquid phase containing A and B will lower the tern- perature a t which B can exist as solid phase, and this lowering will be more than i t would be if A and C were miscible to some extent. When the three components are miscible, the sum of the single depres- sions is usually greater than the depression for the mixture; but this is not always true, owing to disturbing conditions which are not yet defined. H. C. Mass Law Studies, II., 111. By S. F. TAYLOR (J. Pl~ysicccl Chern., 1897, 1, 461-473 and 542-546).--In order to study a case in which two liquid phases and a vapour phase are present, the author has analysed six mixtures of benzene, water, and alcohol.The general form of the relations for this system is deduced from the mass law, and i t is shown that one must use mass concentrations and not volume concentrations in expressing the distribution of a substance between two liquid phases. Agreement of theory and experiment is obtained in the case studied, and the theory is also successfully applied to the system chloroform, water, and acetic acid. H. C. Hydrolytic Dissociation. By HEINRICH LEY (Be?.., 1897, 30, 2192-2196).-The author makes a preliminary communication of results obtained in an investigation of the hydrolysis of salts in aqueous solution.The concentration of the hydrogen ions is deter- mined by measuring the velocity of inversion of cane-sugar by the salt solution a t 100'. The following numbers mere obtained with solutions of aluminium chloride containing 1 gram equivalent in II litres, p being the percentage hydrolysed. t' . P. 32 8.8 64 13.8 128 20.1 Aluminium sulphate is hydrolysed to a smaller extent. Zinc chloride and sulphate exhibit a similar relationship, the latter salt undergoes very little dissociation, the hydrolysis reaching 0.03 per cent. in 1/16 normal solution. Lead and copper chlorides gave abnormal results, the rate of inver- sion of the sugar increasing with the duration of the experiment. Mercuric chloride could not be examined by this method, because it is reduced by sugar. By assuming that its electrical conductivity is entirely due to the hydrochloric acid produced by its hydrolysis, anGENERAL AND PHYSICAL CHEMISTRY. 6’9 upper limit for the latter is obtained varying from 0.29 per cent.in 1/16 normal t o 1.64 per cent. in 11256 normal solution. The addition of potassium chloride t o a solution of aluminium chloride diminishes the hydrolytic dissociation of the latter. The same is true of other salts ; the acid reaction of solutions of lead or mercuric chloride disappears when sodium or potassium chloride is added t o them. T. E. Formation of Anilides. By HEINRICH GOLDSCHMIDT and CURT WACHS (Zeit. physikal. Chem., 189’7, 24, 353-365).-1t has been shown by Qoldschmidt and Reinders that, in solutions of aniline salts in aniline, the salt is‘ probably decomposed into acid and base (Abstr., 1896, ii, 556), a part of the acid being further dissociated.The formation of an anilide is hence analogous to that of an ethereal salt, that is, should be a bimolecular reaction unless a strong acid be added, in which case it becomes monomolecular with velocity proportional to the concentration of the catalysing acid (Abstr., 1896, ii, 638). EX- periments with aniline and acet2ic or propionic acid, and with ortho- toluidine and acetic acid showed the reaction to be bimolecular; on the addition of picric, hydrochloric, or hydrobromic acid, it becomes monomolecular, the velocity being accelerated, and proportional to the concentration of the strong acid if sufficiently great. Ths reaction being bimolecular is not a proof of the non-occurrence of autocatalysis as stated by Donnan (Abstr., 1897, ii, 15) since, if .2: is the quantity of acid converted into anilide, then ( a - x ) is the concentration of the acid, and is also proportional to the concentration of the hydrogen ions, so that the velocity is proportional to (a - x)2.The analogy of these solutions to those of hydrated salts is again noticed, and the view that, in solution, the latter are present as anhydrides thus receives support. L. M. J. Genesis of Dalton’s Atomic Theory. By HEINRICH DEBUS (Zeit. physikcd. Chew,., 1897, 24, 325--352).-The author upholds his pre- viously published views (hbstr., 1896, ii, 639) regarding the origin of Dalton’s theory, that is, that Avogadro’s law was held and employed by Dalton, who definitely states that in 1801 he had a “confused idea ” that all molecules were of equal size ; hence that Mjd = k.He, moreover, made frequent use of this hypothesis; thus the ratio of the atomic weights of oxygen and nitrogen was fixed a t 6/7 from the con- sideration of their densities, whilst, further, of the gases nitric oxide, nitrous oxide, and ‘‘ nitric acid,” the first has the lowest density and therefore consists ‘‘of but two atoms,” and since “nitric acid ” is heavier than nitrous oxide, it follows that, “ a n atom of oxygen is heavier than an atom of nitrogen.” The author poiiits out that this reasoning is entirely dependent on the assumption that the “ atomic ” (that is, molecular) weight is proportional to the density; in other words, that N / d = k . Other similar cases occur, but the apparent failure of this mode of calculation in the case of the hydrocarbons, due to erroneous ideas regarding their composition, ultimately led Dalton to abandon this hypothesis as not of general, although of frequent, application. The second part of the paper is entirely a review,68 ABSTRACTS OF CHEMICAL PAPERS. criticism, and refutation of various passages in the reply of Roscoe and Harden to the previous paper (Zeit. phpikccl. Chem., 22, 241). L. M. J. Lecture Apparatus. By W I L L I m R. E. HODGKINSON (Chem. News, lS97, 76, 152).--Volumerneter. A cylindrical vessel of 100 C.C. capacity provided with a deep and broad, hollow, tubulated stopper is connected at its lower end, by a narrow tube having a stopcock, with a graduated tube. It is filled with the liquid to be used from the zero of thegraduated tube to a mark made on the tube of the stopper. After the liquid has been driven into the graduated tube, the stopper is carefully removed, the substance whose volume is to be measured introduced, the stopper replaced, and the liquid brought back to the mark on the stopper. The volume of the substance can now be read off on the graduated tube. For showing that hot water boils in a closed flask when cold water is poured over it, an improved form of apparatus is described consist- ing of a fractionating flask with a stopcock sealed on the side tube, and a wide thermometer, wedged by a rubber ring in the neck, with its bulb just dipping into the hot water. D. A. L.