to connect the dark and bright rays with the elevations and depressions on the limb of the moon. If these dark rays have any existence, it is difficult to account for them by any known cause. It is easy to see that they cannot be shadows of opaque masses situated anywhere in space, because the shadow of no object situated so near the line joining the observer and the sun could look long and narrow to the observer. Do these rays and striations really exist? We know that such appearances are one of the most common optical illusions when we view a bright object on a dark ground. And, in the various descriptions of them which have come to my knowledge, I have not found a single statement that any observer took the precaution to turn his head in order to see whether these rays of millions of miles in length turned with it. Their existence cannot, therefore, be regarded as proven. The following method is proposed as one that will immediately decide the question of the existence of these rays, and give, at least, a rough approximation to the height of the corona. Erect between the eye of the observer and the sun a series of circular screens subtending angles varying between 32′ and the greatest probable limit of the corona, say 38'. As soon as the total phase commences the observer will stand behind the larger screen and see whether he can make it hide the entire luminosity surrounding the sun, and, if not, will note very carefully the position of any object or luminosity not hidden. He will then make similar observations with the other screens in succession until one is found which shows the corona on all sides at once. The law of diminution of light from the inside to the outside of the annulus can, perhaps, be most accurately determined by exposing a number of equally sensitive photographic plates during widely different intervals, and noting the positions of the circles of equal effect on the different plates. It may be remarked that the height commonly attributed to the corona does not seem compatible with that of an atmosphere surrounding the sun in a state of equilibrium. Though such an atmosphere were of hydrogen, and at a temperature of 100,000° Fahrenheit, its density would be reduced one-half in every 350 miles of height. At a height of 1' in arc, or 26,000 miles, it would be so rare as to be altogether invisible, unless we suppose the temperature to be many times 100,000°. It seems, then, that the general atmosphere of the sun, if there is one, would be altogether hidden by the moon in most total eclipses of the sun, and that its greater brilliancy on the side over which the moon least projected would be very strongly marked. The search for intra-Mercurial planets will be facilitated by the use of similar screens of a larger size. The following proposition is laid down as one hardly admitting of doubt. If the great motion of the perihelion of Mercury deduced by Le Verrier from the observations of transits of that planet is caused by the attraction of matter in the solid or liquid state, that matter can be seen during a total eclipse if the direct light of the corona and protuberances is shut off from the eye. This is founded on the following propositions : The masses of matter in question cannot be more than half of Mercury's distance from the sun. Hence, taking equal surfaces of them and of Mercury, they will shine with more than four times the brilliancy of that planet. If they are not smaller than the second magnitude they will be visible to the naked eye, properly prepared by being kept in previous darkness and shaded by the screen. If they are smaller than the second magnitude, they must be a hundred or more in number. Some of them will then be easily detected with a telescope of large field of view unless they are below the fifth magnitude. If they are below the fifth magnitude, they must number thousands, and they will then be visible to the naked eye as a continuous diffused light. The screen for this search should be large enough to protect the telescope from the sun during the entire period of the total phase. It is hoped that these arrangements and observations will be found to merit the attention of a share of the observers during the total eclipse of August next. SCIENTIFIC INTELLIGENCE. I. PHYSICS AND CHEMISTRY. 1. On the refrangibility of the brilliant yellow ray of the sun's atmosphere. The fact that the yellow ray seen in the spectrum of the protuberances from the sun's photosphere does not correspond with the lines D of the solar spectrum was first observed by Lieut. Herschel, then by Mr. Lockyer, by Secchi, and finally by Janssen. Rayer has now determined the position of this line with accuracy, using a spectroscope of high dispersive power and a plain micrometer. Taking the distance between the two lines D as unity the author found for the distance of the yellow line from the more refrangible ray D the number 2:49, with a probable error of less than 0:03. Taking the wave lengths of the two lines D respectively as 590 53 and 589.88, that of the yellow line is 588-27 in millionths of a millimeter; it corresponds to the division 1016-8 of Kirchhoff's scale.-Comptes Rendus, lxviii, 320, Feb. 1869. W. G. 2. On the presence of the vapor of water in the neighborhood of the solar spots and on the spectral study of certain stars.-By observing the regions adjacent to the large solar spots with a spectroscope of high dispersive power, Secchi has frequently noticed a series of equidistant nebulous lines or bands in the red and orange near the ray 8095 of Kirchhoff and 864 of Kirchhoff. These bands differ in intensity; they appear to consist of very fine rays enveloped in nebulosity and are seen in the penumbras and in the group of small spots recently visible, but usually disappear in the sun's full disc and are wanting in the interior of the large spots where the rays never have the form of the bands. On the 6th January the bands were seen upon the full disc of the sun, but were found to arise from a cirrhus in front of the telescope and disappeared with the cirrhus itself. Secchi remarked that under these circumstances the bands due to the neighborhood of the solar spots were sensibly increased in intensity. By studying the region near D with a spectroscope of nine prisms the author found that the yellow ray of the protuberances really exists in the sun and may be recognized far from the border. The author then sought for the yellow ray in the spectra of the stars which most resemble the sun and found it in Aldebaran, a Orionis and Pollux. Sirius presents a bright region in the corresponding place. In the red stars of the 4th type bright yellow rays may be seen like threads of gold, but it is difficult to fix their positions. Secchi concludes from his observations that the vapor of water exists in the solar atmosphere in the neighborhood of the large spots. The author has also examined the spectrum of Sirius to determine whether there is any displacement of the hydrogen lines due to a proper movement of the star, a question already examined by Mr. Huggins. With a spectroscope of four prisms the ray F was observed to be sensibly displaced, the displacement of the center being sensibly equal to the breadth of the rays D' D" of sodium and being toward the less refrangible side. With a spectroscope of two prisms the displacement of the hydrogen rays a and y with respect to the rays C and F of Sirius was also observed and in the same direction. It is now well known that the bright lines in the nebula are due to hydrogen and nitrogen, but as nitrogen presents spectra of two different orders it was interesting to determine to which of them the bright green line belongs. By direct comparison Secchi found that the line in question corresponds to a brilliant line in the spectrum of the second order. As the production of this line requires a considerable electric force and a higher temperature it is evident that the matter of the nebula is in the more advanced state of dissociation which corresponds to the second spectrum.-Comptes Rendus, lxviii, 358. W. G. 3. Spectral observations of the star R in Gemini.-SECCHI has also examined the spectrum of the variable star R Gemini (R. A. 6h 59m 22s § 22° 54') which star obtained its maximum brightness with a magnitude of 6.5 in February last. The spectrum of this star exhibits a brilliant hydrogen ray (which of the four known to belong to hydrogen is not mentioned). It also presents other luminous bands of which the principal correspond to dark bands in the spectrum of a Orionis intermediate between 6 and D. From this it appears that the spectrum of this star is analogous to that of the variable star in Corona Borealis which appeared in 1866.- Comptes Rendus, lxviii, 361. W. G. 4. On absorption lines produced by the passage of the solar light through chlorine.-MORREN has found that by employing a spectroscope of five prisms of highly dispersive flint glass, absorption lines are distinctly visible in the spectrum of light which has traversed a tube filled with chlorine two meters in length. The lines begin to be visible in the part of the spectrum near b. They vary in intensity, fineness and mode of grouping and exhibit some slight free spaces. They have no regular order and extend beyond the ray F toward the ray 2110 of Kirchhoff's scale. In this last portion they are very numerous and almost equidistant. The solar spectrum proper continues visible as far as 2210, but after that the light is completely absorbed. Chlorine therefore absorbs the colored portion of the spectrum where the chemical rays are most abundant.-Comptes Rendus, lxviii, 376. W. G. 5. On hydrogen in its relations to palladium.-GRAHAM has shown that palladium absorbs 800 or 900 times its own volume of hydrogen, its density being sensibly diminished. When a wire of palladium is charged with hydrogen by being made the negative pole of a battery, it increases in length and volume. By measuring this increase and determining the volume and weight of hydrogen absorbed, Graham found for the constituent of the alloy, palladium 95.32, hydrogen 4.68, and for the density of the hydrogen compressed by occlusion in the metal numbers varying from 1708 to 2055, the mean of the numbers accepted being 1.951. The author considers the hydrogen under these circumstances to be in the metallic state and calls it hydrogenium. The tenacity and electric conductibility of palladium are both diminished by the absorption of hydrogen, the former in the ratio of 100 to 81.29; the latter in the ratio of 8.10 to 5.99; on the other hand the paramagnetic character of the metal is decidedly increased by its association with hydrogen which must be classed among the magnetic elements with iron, nickel, cobalt, chromium and manganese. At a high temperature hydrogen passes with considerable facility through palladium. The greatest rate of passage observed was four liters of gas per minute through a plate of palladium one millimeter in thickness and one square meter of surface, the temperature being a little below the melting point of gold. From the above it appears that hydrogenium is a solid white metal of density about 2, magnetic, conducting electricity and uniting with palladium in the proportion of one atom to one atom. The alloy possesses the properties of gaseous hydrogen intensified. It reduces mercury and calomel from mercuric chlorid and possesses in general a considerable degree of reducing power.-Comptes Rendus, lxviii, 101. W. G. 6. Production of an artificial spectrum with a single Fraunhofer's line. When the discharges of a Holtz machine aided by a Leyden jar of about one foot of internal coated surface and short striking distance are made to pass through a Geissler tube placed in front of the slit of a spectroscope, the spectrum of the gas in the tube is first seen. Wüllner finds that if the striking distance be slightly increased the sodium line makes its appearance, and with an appropriate striking distance is so bright as to far exceed in intensity the lines of the gas-spectrum. When the striking distance is again slightly increased the bright lines of the calcium spectrum make their appearance with a beauty and sharpness which can scarcely be attained in any other way. By again increasing the striking distance the line of light in the tube becomes extremely brilliant and then exhibits a bright continuous spectrum in which, however, the sodium line appears perfectly dark. The explanation of this phenomenon is found in the fact that fine splinters of glass are torn from the inner side of the tube. These become intensely ignited and their light gives the continuous spectrum. The ignition of these particles takes place in an atmosphere of the vapor of sodium from the glass so that the same current produces the ignited core and the absorbing atmosphere. As the calcium lines are not seen inverted it is probable that the density of the calcium atmosphere is not sufficient to produce inversion.-Pogg. Ann., cxxxv, 174. W. G. 7. On the analysis of different varieties of carbon.—Berthelot has communicated to the Academy of Sciences a very elaborate and detailed memoir on the different varieties of carbon, showing that the number of modifications of this element is much greater than had hitherto been supposed and, as seems to us at least, full of interest. The author in the first place recalls the three recognized varieties under the heads of diamond, amorphous carbons derived from organic matter, and graphites, natural and artificial. The method of study employed was based upon Brodie's method of oxydizing graphites by means of nitric acid and potassic chlorate. By these reagents diamond is not sensibly oxydized; the different varieties of amorphous carbon are changed into humus like substances of a yellowish brown color soluble in water and varying according to the variety of carbon analyzed, while the graphites are converted into graphitic oxyds which differ with the nature of the graphites which furnish them. Berthelot gives the name pyrographitic oxyds to the black powders which remain when the graphitic oxyds are heated. They also contain oxygen and hydrogen as well as carbon. Native plumbago, or as Berthelot terms it, plumbagine, already examined by Brodie, yields a graphitic oxyd |