General and Physical Chemistry. Dissociation of Optically Active Salts in Solution. By PH~L~PPE A. GUYE and B. ROSS (Bull. Xoc. Chim., 18!J5, [3], 13, 464-469).-1t is well known that the specific rotatory powers of the salts of optically active monobasic acids, at sufficiently low concentra- tions, are identical with those of the acids themselves, whilst those calculated from the rotatory powers of more concentrated solutions are usually divergent. Although this is accounted for in a general manner by the theory of electrolytic dissociation, the optically active acids being, as a rule, comparatively little dissociated, and having approximately normal cryoscopic constants and electrolytic conduc- tivities at temperatures at which their salts are almost completely dissociated ; yet a complete examination of an individual case has not hitherto been made.The authors have, therefore, determined the rotatory powers of various inorganic and organic salts of active valeric acid, a substance which was chosen on account of the comparative simplicity of its constitution due t o the absence of alcoholic hydroxyl. The specific rotatory power of the aqueous solution of the acid is [:ID = + 17*3O, whilst that of the pure acid is +13*64O; but this difference may be due as much tso molecular association as to electro- lytic dissociation. The specific rotatory powers of the lithium, sodium, potassium, and rubidium salts at concentrations corresponding with 2.46 grams of the acid i n 100 c.c., and calculated on the amounts of the salts actually present, are respectively [a]D = +€POo, + 7.2O, +6*4O, and +5.4' ; whilst those calculated on the amounts of acid present, on the sup- position that the salts are conipletely dissociated into their ions, are respectively + 8*5O, + 9*8O, + 8*9", and + 10.1O.The approximate identity of the rotatory powers of the salts of metals differing so widely in atomic weight, is noteworthy, considering the great dif- ference between those of the alkylic valerates ; the lithium, potassium, and rubidium salts are the most dissociated, the mean rotatory power of the three being approximately half that of the acid at the same concentration. As might be expected from theory, the specific rotatory powers of aqueous solutions of valeric acid me diminished alike by dilutiori and by rise of temperature. The rotatory power, [aJD = + 12*02O, of an acid somewhat less active than that used in the foregoing determinations, became +14.6" in an aqueous solution containing 3-732 grams per litre, but was diminished to +14*4', when the concentration was reduced to 1-239 gram per litre, and was rednced in the two cases to +14*36O and +14*2', when the temperature was raised from 18" to 25O. The rotatory powers of aniline, pyridine, and diethylamine valerates in aqueous solutions of the same concentration of acid as those of the alkali salts, and calculated on the amounts of the salts actually present, are [a]= = +6*30°, +6*36O, and +4*99O respectively, whilst YOL.LXX. ii. 886 ABSTRACTS OF CHEMICAL PAPERS. those calculated on the amounts present on the hypothesis of complete dissociation are + L2.4', +11.26', and +8*77O.These results are in harmony with the cryoscopic constants of the bases. The specific rotatory powers of alcoholic solutions of the foregoing salts and of those of dimethy laniline, quinoline, and phenylhydraaine containing the same amounts of acid as before, are practically identical with that of an alcoholic solution of the acid of the same concentration, except in the case of the diethylamine salt, the activity of which is some 50 per cent. less. With this exceptioc,. therefore, these salts are completely dissociated in alcoholic soluticin, a conclusion similai. to that arrived a t by Ghira from a study of the cryoscopic properties of benzene soliitiona of the corresponding acet'ates.It is thus useless to determine the optical activity of solutions of the organic salts of optically active substances in organic solvents ; and where organic solvents are used, investigation must, in future, be limited to inorganic salts. Colour Change of Dilute Solutions of Potassium Chromox- alate. By FRIEDRICH HAMBURGER (Ann. Phys. Cheni., 1895, [el, 56, 173--174).-A dilute solutiofi of potassium chromosalate, K3Cr(C2O&, when placed in a cylindrical vessel was found to exhibit in general a green colour by transmitted daylight, hut showed purple streaks a t the edges and in the centre. By artificial light, the solution appesred only purple. An examination of the absorption spectrum was there- fore made, and showed a, wide absorption band between the wave- lengths 630 and 530, and complete absorption at the wave-length 4iO.Hence the yellow, orange, blue, and violet rays are absorbed from the spectrum, only red and iudigo remaining ; the colour resulting from these two is dependent on the proportion in which they are present in the incident light rays. Emission of Light by Organic Substances in the Gaseous, Liquid, and Solid Condition. By EILHARD WIEDEMANN and G. C. SCHMIDT (Ann. Plys. Chew., 1895, [2], 56, 18--26).-The authors find by numerous experiments that many organic compounds exhibit fluorescence in the gaseous condition; this fluorescence is blue or violet with retene, phenanthreae, anthracene, anthraquinone, chrjsene, indigo, and naphthalene, and a magnificent reddish-brown with napht.hazarin. 1 he absorption spectra of solutions of the hydro- carbons in this list, all lie in the extreme violet or ultra-violet, and, therefore, a s in the case of benzene and toluene, the absorption of the gaseous hydrocarbons probably lies still further towards the ultra- violet end of the spectrum. Anthrnquinone also absorbs violet rags, and nsphthamrin, i n alcoholic solution, green, blue, and violet rays.The emission spectra lie, therefore, in accordance with Stokes' rule, nearer the less refrangible end of the spectrum than the absorption spectra, and the spectra of organic vapours are fluorescence spectra in the nsnal sense of the term. Many organic compounds in the gaseous state yield characteriatic spectra under tahe influence of electrical discharges, but the spectra, obtained in this way do not correspond with the absorption spectra.In JN. W. H. C.GENERAL AND PECYSICAL CHEMISTRY. 57 the liquid state many organic compounds become luminous under the influence of the cathode discharge, and the colour in this case is the same a s that of the vapour. Solid organic compounds also frequently exhibit cathode-luminosity, but the colour is not in all cases identical with that of the liquid. By L. HOLBORN and W. WIEN (Ann. Phys. Chem., 1895, [2], 56, 360-396).-The Le Chatelier thermo-couple of platinum and a platinum-rhodium alloy containing 10 per cent. of rhodium, has been compared by the authors with the air thermometer up to the temperature 1450'. This instru- ment gives very constant readings i f protected from the possible action of carbon.Its use in high temperature measurements is, there- fore, to be preferred to that of employing the change of resistance of a platinurn wire, for with the latter, even if between each measure- ment the temperature coefficient between 0" and 100' is redetermined, this affords no guarantee for the behaviour at high temperatures. The authors find, also, that Callendar's formula for the change of resistance with change of temperature is not sufficiently exact to admit of a far reaching extrapolation. A comparison with the air thermometer can be most readily effected with the thermo-couple, as it can be directly introduced into the air vessel, and must in this way acquire the same temperature as that of the expanding air. The following melting points were measured :- H.C. Measurement of High Temperatures. Silver.. .......... 971' . Nickel. .......... 1484' Gold ............ 1072 Palladium.. ...... 1587 Copper. .......... 1082 Platinum . . . . . . 1780 H. C. i Some Melting and Boiling Points. By HENRI LE CHATELIER (Compt. reihd., 1895, 121, 32&-326).-Nxperiments made with ft thermoelectric couple protected by a thin film of glass and graduated up to the boiliug point of zinc (930') give 1050'to 1060' as the melting point of gold. Experiments made with another couple, taking the melting point of silver as 954', show that the meltiug point of gold is 1O55-1O6O0. The author concludes that the melting point of gold as given by Violle, 1045O, is too lorn, but that the error is certainly noh more than 20°, that none of the more recent estimations of t h i s melting point are sufficiently accurate t o warrant their sabstitution for Violle's number, and that it is very desirable to retain the scale of high tern- peratureR at present actually in use until new and more exact experi- ments made by direct comparison with the air thermometer shall give the true melting point of gold to within a very few degrees (compare Heycock and Neville, Trans., 1895, 1064).Explosion of Endothermic Gases. By LEON MAQUENNE (COW@. rend., 1895, 121, 4 2 G 4 2 i ) .-When a small quantity of mercuric fulminate is exploded in contact with nitrous oxide, the latter is de- compolred with a violent explosion. Acetylene under similar condi- tions only begins to decompose, about 95 per cent. of the gas remaining unaltered, and the explosive wave is not propagat,ed through the masB of the gas.With a larger quantity of fulminate C. H. B. 8-218 ABSTRACTS OF CHEMICAL PAPERS. (about 1 gram) the acetylene is decomposed, and the explosive wave travels through a distance which depends on the diameter of the tubes and the conditions under which the explosion takes place. Nitro-substitutions. By CAMILLE MATIGSON and DELIGNP (Conzpt- C. H. B. rend., 1895, 121, 422-424).- Ecat of cornbustion. Const. volume. Orthonitrophenol . . . . . . 688.6 Paranitrophenol ... .... 689.5 Orthonitrobenzoic acid . 731.1 Metanitrobenzoic acid.. 727.7 Paranitrobenzoic acid . . 729.6 Paranitroacetanilide.. . . 969.2 Nitrobenzaldehyde . . . . . 801.2 I Const. pressure. 688.2 689.1 730.4 727-0 728.8 968.9 800.3 Difference from parent compound.44.3 43.4 44.0 47.4 45.5 47.9 41.4 The third column gives the difference between the heats of com- bustion, at constant pressure, of the nitro-derivative and the parent substance. The position-isomerides have practically the same heats of combustion. With the exception of nitrobemaldehyde, the difference is practically constant and is about 45, whatever the function of the compound in which substitution takes place. The equation deduced from this difference is ihe thermal disturbance being indentical with that found by Berthelot in the case of the hydrocarbons benzene, toluene, naph- thalene, &c. C. H. E. Combination of Mercuric Cyanide with Bromides. By RAOCL VARET (Compt. rend., 1895, 121, 398400).-In the following table, column I, gives the heat of dissolution of the salt, column I1 the heat developed by the interaction of solutions of mercuric cyanide and the particular bromide, aud column TI1 the heat of formation of the solid salt from its proximate constituents, the salts being regarded as solid and the water as liquid. RCH + HNO, liq.= RCNO, + H20 liq. develops +36*7 Cnl. 2Hg( CN)2,2NaBr,4Hz0.. . 2Hg(CN),,2LiBr,7H20 . . . 2Hg(CN)z,BaBr,,7Hz0 . . . ‘LHg(CN)2,SrBrz,6H,0 . . . 2Hg(CN)2,CnBrz,7H20 . . 2Hg(CN)2,ZnBrz,8H,0. . . Hg(CN)2,CdBr,,3H20.. . . 2 Hg (CN) 2,2NH4Br,2 H2O. . 28g(CN)2,MgR~~,8€320. . . I. -20.97 ,, -18.31 ,, -20.98 ,, -18.60 ,, -19.82 ,, -15.97 ,, -20.82 ,, -12.5 ?, -24.14 Cal. IT. 111. f0.98 Ca.1. + 18.52 Cal. +125 ,, +36*26 ,, +1.29 ,, +20.27 ,, +1.24 ,, +29*84 ..f l . 2 5 ,, +40*47 ,, +1*44 ,, +54*71 .. +133 ,, +31.15 ,, +0%6 ,, +10.5 ?, +1*06 ,, + 7.23 ,, At the ordinary temperature, the solutions are slightly alkaline to litmus, and slowly give the isopurpurate reaction with a picrate of the same base as is combined with the bromine. It follows that whilst; almost all the mercury is in combination with cyanogen, aGENERAL AND PHYSICAL CHEMISTRY. 89 small quantity is in combination with the bi-omine. of the latter increases as the temperature rises. accord with the thermochemicnl data. The proportion These result,s are in C. H. B. Distillations with an Automatic Mercury Pump. By FRIEDRICH KRAFFT and W. A. DYES (Ber., 1895, 28, 2583-2589).- The paper contains a description of an air pump devised some 16 years ago by v.Babo. It consists of a Sprengel mercury pump which works in the vacuum of a water suction-pump and so need not be more than 30-40 cm. long ; the mercury, which falls down the short Sprengel tube, is carried up again to the top of the apparatus by a current of air drawn in by the water-pump, aud the apparatus is thus automatic and continuous in its action. It is comparatively small in size, so that it can be placed, by moans of clamps and a retort-stand, on the laboratory bench, and it does not require more than 600-650 grams of mercury ; a vacuum of less than 1 inm. is attainable by its means. By means of this pump the boiling- or sublimation-points of several substances were determined. Manni to1 boils at. 276-280' under about 1 mm., at 285' under 2.5 mm., and at 290-293' under 3-3.5 mm.pressure. Dulcitol boils at 275-280', 287-288', and 290-295' under the same pressures. a-Hydroxybutyric acid boils at 84" under 1.5 mm. pressure. Succinic acid sublimes at 156-157' under 2.2 mm., at 160-165' under 2.5-3 mm. pressure ; fumaric acid at 163' under 1.7 mm., mesaconic acid at 139-141' under 1.5 mm., at 14.3-145' under 2 mm., and itaconic acid at 140-141' under 1-5 mm. pressure ; mnleic and citraconic acids do not sublime without forming some anhydride. Wood's metal was used to heat the distillation flasks; the temperature of the bath need not be more than a few degrees above the boiling-point oE the liquid, but if the substance does not boil, but sublimes, a much greater difference of temperature, 40-60' above the sublimation point, was found to be necessary.Density Determinations of Extremely Dilute Solutions. By FRIEDRICH KOHLRAUSCH (Ann. Phys. C'hem., 1895, [S], 56, 185-200). --The author has further improved the method of determining the density of very dilute solutions, which he had formerly employed in conjunction with Hallwachs (Abstr., 1894, ii, 441). The size of the ball of glass used was greatly increased, the suspension from the balance being effected by means of a thin platinum wire which was first coated electrolytically with platinum and afterwards ignited. The rough surface thus produced, satisfactorily removes the difficul- ties and irregularities observed when a smooth wire is employed. As very slight variations of temperature exercise a very disturbing influence on the results, owing to the unequal expansion of the soh- tion and of the glass ball, it was necessary to confine the observations within limits of temperature between which this inflnence is at a minimum, the limits within which a variation of a few thousandths of a degree may be allowed, being from 4' to 8'.A limiting error of lo-' has thus been reached in the determinations. Observations were made with solutions of cane sugar, magnesium C. F. B.90 ABSTRACTS OF UHEMICAL PAPERS. phste. A = 60.23. I A = 60'00. sulphate, acetic acid, and sulphuric acid. On the assumption that the water undergoes no contraction in volume, the molecular volumes @ of the dissolved substances are calculated by means of the formula s - 1 m c) = A - 1000 - , where A is the weight of the equivalent, m the number of gram- equivalents per libre in the solution, and s is the density. The resnlts are given in the following table, arid for the purpose of com- parison, t.he results formerly obtained with Hallwachs a.t the higher temperature of 18" are included.A = 49 -04. - m. -- 0 -0002 0 -0006 0 -001 0 *002 0.005 0 *01 0 *03 0.05 0.1 1 5 - - - - 50.7 50.88 51.0 51-04 51.10 51.34 52.14 Sugar. A = 341 '1. (6 '1) ( 5 . 5 ) (65:;) 6.92 7.71 9.75 10.75 12-03 I 15.54 17-57 6 '0". 207 '0 207 *3 207 -32 207 *41 207 48 207 *56 207 -70 207 -8 208 -0 203 *9 - 18". - - 209 -0 209 *O 209 -5 209 -69 299 -71 209 -57 209 -89 211 -5' 215 -9 6 *3". -4.5 -4.5 -4.6 -4.6 -4'14 -3 *91 -3 '37 -3 -03 '-2 -45 + 0 . 9 + 6 *o - - - -3 *4 - 3 '21 - 2 '65 -2 a15 -1.74 -1 *21 + 1 *68 + 6 *58 (51 '3) (49 '8) (50 -0) 49 -61 49 *69 49 '72 49 -85 49 '88 49 '93 50 *21 51 *05 18". -- - -.- 6 -9 7 '94 9.32 11 -80 12 -77 14 '05 16 -96 18-52 It will be seen that the continual increase in the molecular volumes with rising concentration is to be observed in these as in the former determinations. The only exceptions occur in very dilute solutions of acetic and sulphuric acids; here a complicated behaviour or disturbance occurs similar to that noticed in couduc- tivity determinations, the molecular contraction increasing only up to a certain degree of dilution, and then decreasing. It is probable that the explanation of this peculiarity may be found in the traces of impurities present in the water used, as these undoubtedly affect the density determinations in these extremely dilute solutions in a marked manner.The molecula,r volumes are throughout one to two units greater at 18' than at 6". H. C. Influence of Hydrochloric acid and Chlorides on the Photo- chemical Decomposition of Chlorine Water. By EUTHYM E KLIMENKO (Ber., 1895, 28, 2558-2564).-Norrnal solutions of hydro- chloric acid and of various chlorides were diluted with equal volumes of chlorine water and exposed in sealed tubes to sunlight, together with some tubes containing chlorine water only. When the chlorine in the latter had completely disappeared, the other tubes wereQENEHXL AND PHYSICAL, CHEMISTRY. 91 opened, and the amount of chlorine still remaining in them was estimated. It was found that the hyclrocliloric acid had most retarded the disappearance of the chlorine, very little chlorine having disappeared in this case ; if the amount left in the hydrochloric acid tube be taken as 1, then the amounts left in the tubes containing the chlorides aye represented by the following numbers : LiCl 0.308, NaCl 0.173, KCl 0.090, MgCI, 0.530, CaCl, 0.390, SrC1, 0.302, BnCI, 0.285, ZnC1, 0.200, CdCI, 0.042.The different metals group themselves in the same order as in the periodic system. The action of chlorine on water is said to result in the formation, first of hydrochloric and hypochlorous, and finally of hydrochloric and chloric, acids. The retarding action of hydrochloric acid is ascribed to its reconverting the hypochlorous acid formed into chlo- rine. It is further suggested that metallic chlorides are partly con- verted into chlorates by the chloric acid formed, and that the hydro- chloric acid thus liberated reacts with hypochlorous acid to reform chlorine, owing t o which circumstance the metallic chlorides retard the disappearance of the chlorine.C. F. B. Heating Apparatus for Drying Ovens. By JOHANNES THJELE (Uer., 1895, 28, 2601--260'L).-The tubes which carry the rows of small jets used in heating drjing ovens are usually clamped to the legs of the stand by means of screws ; these screws become so hot that it is generally impossible to touch them, and consequently diffi- cult to regulate the temperature of the oven. The author has devised a stand, figured in the paper, in which the tube of jets is carried by a rail that slides up and down between two of the legs, and is pressed against these by a spring, so that it is held fixed in any position. To alter the position, it is merely necessary to press two handles together ; these act as levers, and release the pressure of the spring ; they are, moreover, fairly long and have wooden ends, so that they do not get too hot to touch. . A Modified Condenser. By J. J . L. VAN RJJN (Ber., 1895, 28, 2388).-The modification consists in bending the inner tube so that the whole condenser can be rotated, without being removed from the vessel to which it is attached, in such a manner that it can be used eifher as an ordinary condenser or it reflux condenser. It appears from the drawing which is appended, that the modified condenser could not be used in connection with any vessel with a, narrow aperture. A. H. C. F. B. Modification of Liebig's Condenser. By HUGO MICHAELIS (Bey., 1895, 28,2615).-The author brought forward the modification recently described by van Rijn (preceding abstract) about 10 years a.go (Chem. Zeit., 1886, 1556). A. H.