of the rocksalt; still it is possible that fluorspar at 150° emits rays other than those which it absorbs at ordinary temperatures. This behavior is however probably connected with the great reflecting power of fluorspar for rocksalt-heat, of which I shall speak in the second part of this memoir. (9.) If it were possible to produce a spectrum of the heat radiated at 150° C., the spectrum would, if rocksalt were the radiating body, exhibit only one luminous band. If sylvin were used as a radiator, the spectrum would be much more extended, but would still occupy but a small portion of the spectrum which the heat radiated from lamp-black would form. In concluding this part of his memoir Magnus makes some remarks on transparency which seem to us very suggestive. If we assume that there is a constant interchange of heat even between bodies having the same temperature, we may fairly assume also that there is such an exchange in the case of light. We cannot observe the light which bodies emit at ordinary temperatures, but they do not absorb light, since this absorption produces their colors. If such an exchange of light takes place at ordinary temperatures, it would follow that transparent bodies either radiate only such rays as are not contained in the light emitted by ignited bodies, and then they absorb none of these rays, or that they emit only one or a few of the wave-lengths of the light which is visible to us; since then they absorb only these, and allow all others to pass through, so that the intensity of the transmitted light is but little less than that of the incident light. We may therefore infer that the transparency of bodies depends upon the fact that they only radiate a few of the wave-lengths, which are contained in the light known to us.-Pogg. Ann., Band cxxxix, p. 431. W. G. 2. On the breadth of the spectral lines.-LIPPICH has given an explanation of the breadth of the spectral lines of luminous gases, which depends upon the now generally received dynamical theory of gases and vapors. The explanation amounts to a mathematical theory of the broadening of the bands, the diminutions in their intensity, the addition of new bands, and the appearance of parts of continuous spectra, all of which appear to depend upon changes of temperature. In the present paper the author treats only of the application of the theory to the widening of the bands. The fundamental assumption is this; if it be necessary to consider a molecule of a gas as a system capable of vibration, the spectrum of an ideal gas in which the molecules would be perfectly free elastic systems, could consist only of a number of different colored bands of absolutely homogeneous light, if the vibratory motions of the molecules alone are taken into account. According to the theory of Krönig and Clausius, the molecules of a gas have progressive motions, with very great velocities, and this circumstance in connection with the well known influence of the motion of a luminous point upon the refrangibility of the emitted rays, makes it possible to explain the widening of the spectral bands, and to show the dependence of the breadth of the brands upon the temperature and density of the ignited gas. Setting out with these principles, the author arrives at the following law; The ratio of the difference of the wave-lengths which correspond to the borders of a spectral band, to the mean wave-length of this band, is for one and the same gas constant for all the spectral bands, and in different gases is directly proportional to the square root of the absolute temperature, but inversely proportional to the square root of the density. 1 The breadths of the bands, as they appear in the spectroscope to the eye, will increase with the refrangibility of the rays somewhat more rapidly than The author in the next place shows that in the case of dark bands in a spectrum, produced by inversion, the distribution of the darkness, so to speak, and the breadths of the band will follow the same law as that above given for bright bands. The mere comparison of the relative breadths of the bands may lead to important conclusions. If in any gas-spectrum bands of different breadths lie near together, we might infer the presence of a mixture of gases of different densities, or of different allotropic states of the same gas. Thus the author thinks that the presence of some fine lines in the blue of the oxygen-spectrum may indicate the presence of the denser ozone. Furthermore, the observation of the breadths of these lines may decide whether the appearance of new lines is due to the more intense vibrations of the same molecules, or to their allotropic conditions. For the same gas the breadth of the spectral bands permits a conclusion as to the temperature. This is of special interest in the case of the heavenly bodies, in the spectra of which a widening especially of the hydrogen lines has been observed. The author does not agree with Frankland and Lockyer in ascribing this result solely to the higher pressure of the gas, but thinks that it must also have a much higher temperature. As Huggins and Lockyer have observed a widening of certain dark lines in the spectrum of the umbra of a solar spot, the author infers that the gases and vapors forming the umbra must have a higher temperature than that of the part of the sun's atmosphere, in which Fraunhofer's lines originate. Finally, Lippich suggests that measurements of the breadths of the spectral bands will finally lead not merely to relative but to approximate absolute values of temperature and density. Among other results, it may be possible in this way to furnish an experimental proof of the correctness of the dynamical theory of gases. In conclusion, the author remarks that, strictly speaking, the results which he has obtained apply only to perfect gases. Changes in the spectrum will indicate changes from gases to vapors. New periods of vibration will occur, which will exhibit themselves in the spectrum with the less intensity, the greater their deviation from the periods of vibration of the molecules of the perfect gas. In this manner a spectral band will fade out on both sides, and with increased pressure the illuminated portion will become broader the more the gas deviates from the law of Gay Lussac and Mariotte. This explanation agrees with Wüllner's experiments, according to which the "washing out" of the bands occurs in the cases of oxygen and nitrogen, under much lower pressures than in the case of hydrogen.-Pogg. Ann., cxxxix, p. 465. W. G. 3. On the position of thallium among the elements.—RAMMELSBERG has contributed some interesting facts in relation to the compounds of thallium, without however furnishing data for determining the degree of atomicity of the metal. The author, in the first place, calls attention to the fact that the isomorphism of thallium in its compounds with potassium and sodium is not sufficient to decide the question. When thallic sesquioxyd is heated with a solution of iodic acid, a normal dithallic salt is formed, which has the formula TI(I)+3aq. The sesquioxyd dissolves easily and completely in chlorhydric acid, forming a colorless solution, which after addition of potassic or ammonic chlorid yields beautiful crystals. The new salts, respectively TICI,,6KCl+4aq. and TICI,6NH Cl+4aq., form large colorless transparent crystals, which resemble combinations of the cube, octahedron and dodecahedron, but which really belong to the square prismatic system. They are not decomposed by water, even on boiling. With bromine and potassic bromid, and iodine and potassic iodid, thallium forms the salts, TIBг6,3KBr+3aq. and FII,3KI+3aq., which crystallize in regular octahedra. Rammelsberg remarks that the thallium atom, Tl=204, can hardly be regarded as other than monatomic, but as the double atom of the molecule in the dithallic compounds is hexatomic, the single atom would have to be considered as tetratomic. The specific heat of thallium is an evidence that the metal is Tl-204, but the isomorphism of two monatomic with one diatomic atom has been established in so many cases that no certain conclusion can be drawn from the isomorphism of thallium with potassium and sodium. It seems probable that the question can only be settled by determinations of vapor density. Berichte der Deutschen Chemischen Gesellschaft, Jahrgang, 3, No. 7, p. 360. W. G. 4. On a new method for the volumetric estimation of copper.— WEIL has given a new method for the determination of copper which appears deserving of attention. It is based upon the following facts. At a boiling heat and in presence of an excess of free chlorhydric acid the least trace of cupric chlorid communicates a very distinct greenish yellow tint to the solution. This tint is the more intense the greater the quantity of acid present. Stannous chlorid instantly reduces under these circumstances cupric chlorid to colorless soluble cuprous chlorid. The reaction is finished when the green solution of cupric chlorid is completely decolorized. A single drop of stannous chlorid in excess is easily detected by a drop of mercuric chlorid, which gives a precipitate of calomel. When the solution contains iron in the form of sesquioxyd as well as copper, the volume of the solution of tin employed will indicate the sum of the copper and iron. In this case the author precipitates the copper in another portion of the assay by means of zinc coupled with platinum, and then determines the iron by means of potassic hypermanganate. The copper is then easily found by difference. The author determines the litre of his solution of tin by means of pure metallic copper, and preserves it from oxydation under a layer of petroleIt is of course necessary that the solution of copper should be perfectly free from nitric acid.—Comptes Rendus, lxx, p. 997. um. W. G. 5. On the utilization of the secondary products obtained in the manufacture of chloral.-Dr. A. W. HOFMANN has examined a mixture of secondary products obtained during the manufacture of chloral, and condensed during cold weather. The liquid began to boil at 17°-18°, rising slowly to 30°-31°, where the temperature remained constant a short time, and then rising again to 50°, when nearly all distilled over. The most volatile portions were mixed with three times their volume of alcohol saturated at 0° with ammonia and heated in a water bath for an hour. The liquid was then filtered to separate crystals of sal-ammoniac and the alcohol ammonia and chlorinated ethylic chlorids distilled off. The mass of chlorhydrates of ethyl-ammonias remaining were decomposed with caustic soda, and the separated liquid alkalies dehydrated by caustic soda, and finally distilled. In this manner 5 litres of the secondary products operatedĝon gave 1 liters of a mixture of anhydrous ethylamines. These could be separated from each other by means of oxalic ether, in the manner already pointed out by Hofmann. The results of this investigation are interesting, from the prospect which they afford of obtaining the ethyl-ammonias as an article of commerce, at a reasonable price, and in comparative abundance. Comptes Rendus, lxx, p. 906. W. G. 6. On the nature of the secondary products obtained in the manufacture of chloral.-KRÄMER has studied the other products of the action of chlorine upon alcohol, the existence of a large quantity of ethylic chlorid having been shown by Hofmann. As the ethylic chlorid was in contact with an excess of chlorine, it was natural to expect to find in the less volatile oily products the whole series of chlorinated ethylic chlorids described by Regnault, and experiment showed that several of these substances were pres ent. The most volatile product boiling at 60°, proved to be chlorinated chlorethyl or chlorethyliden, CH,Cl,, identical with the chlorethyliden prepared from aldehyd. A liquid boiling at 85°, proved to be ethylen-dichlorid, the formation of which by the action of chlorine upon ethylic chlorid had not before been observed. The next product was chlorinated ethylen dichlorid €2H,Cl. Cl2 boiling at 115°, and the last bichlorinated ethylen, €2Ĥ2Čl2, boiling at 37°. Other chlorinated products were also observed, but not yet studied. To prove the identity of the chlorethyliden obtained in this manner with that obtained from aldehyd by the action of phosphoric pentachlorid, Krämer heated a portion of it. 2 3 8 1 with alcoholic ammonia to 160° for 12 hours. In this manner an oily base boiling at 180°-182°, and having the characteristic odor of collidine, €,H,,N, was obtained. This base had already been formed from aldehyd-ammonia by Bayer, and found to be identical with that obtained by Anderson from animal oil.-Berichte der Deutschen Chemischen Gesellschaft, Jahrgang, 3, p. 257–262. W. G. 7. On some properties of iron precipitated by the galvanic current.-LENZ has given some interesting particulars in relation to the composition and properties of iron as precipitated in the metallic form by the battery. The iron examined was deposited by Klein's process from a solution of the mixed sulphates of iron and magnesium. Weak currents were employed, and the solution was kept neutral by carbonate of magnesia. Iron so thrown down has a beautiful fine-granular structure, showing no traces of crystals under the microscope. Its color is a soft bright gray. Its hardness is very remarkable-not less than 55 of the ordinary mineral scale, and it is excessively brittle, so that it may be rubbed to powder between the fingers. When the iron is slowly reduced upon a polished surface, it is free from flaws and has a velvety look. As it becomes thicker, bubbles or pits are formed as small oval depressions. When heated over a fire the iron loses many of these properties in a remarkable degree. Its hardness diminishes and becomes 45; its brittleness entirely disappears, and it becomes so flexible and tenacious that it cannot be broken by repeated bending or even by folding and strongly smoothing down the folds. When heated in vacuo the iron changes color and becomes almost as white as worked platinum. The ignited iron rusts very quickly both in air and in previously boiled water; this is not the case with the metal before ignition. In the electric series unignited iron stands nearer to copper than the ignited metal. On analysis by means of Sprengel's pump, the precipitated iron was found to contain various gases-vapor of water, nitrogen, carbonic oxyd, carbonic acid and hydrogen. Levy thinks that the carbonic acid came from the solution from which the iron was precipitated, and that the carbonic oxyd was formed during the ignition of the tube containing the iron in the process of analysis; also that the vapor of water was formed by the union of hydrogen with the oxygen of a small amount of rust in the iron, since it was only given off at 1,600° C. The hydrogen in the iron was always in largest quantity; the whole quantity of gas varied greatly, and sometimes amounted to 185 times the volume of the iron. absorption of the gases was found to take place mainly in the first layers formed. On warming the reduced iron, the evolution of gas began at temperatures below 100°, but at this temperature chiefly hydrogen was evolved. Ignited galvanically reduced iron decomposes water and absorbs the free hydrogen either wholly or partially.-Bulletin de l'Académie de Pétersbourg, xiv, p. 337, cited in Dingler's Polytechnisches Journal, cxcvi, p. 44. The W. G. |