to the spectra of the sun and stars, or a spectrum consisting of one, two, or three bright lines, indicating the gaseous nature of their composition. It has been found that when the light of a nebula is dispersed by the prism, the continuous spectrum is very faint, its light being of all refrangibilities. In consequence of this, observers have not yet been able to discover whether it is crossed by dark lines or not. Mr. Huggins' paper contains a list of six of the small planetary nebulæ, in which the spectrum of three bright lines is visible. The Dumb-bell nebula (733, Plate XXX., fig. 3), and one in Lyra give a gaseous spectrum. Mr. Huggins observes that, "since the light from these nebulæ emanates from a gaseous source, we have an explanation of the small intensity of their light; and, it may be, also to some extent of the strange appearances which some of them present, for on account of the absorption by the portions of gas nearest to us of the light from the gas behind them, there would be presented to us little more than a luminous surface." The great nebula in Orion (736) also belongs to this class of gaseous bodies, as shown by its spectrum of three bright lines.

With respect to the probable relation of the gaseous nebulæ to the other nebulæ and clusters, Mr. Huggins considers that a more intense heat may be indicated by the superior intensity of the light of the gaseous nebulæ. It is possible, therefore, that of all the objects usually included among the nebulæ, those which give a gaseous spectrum are, as a class, to be considered as generating more heat than those giving a continuous spectrum.

Since the celebrated researches of MM. Kirchhoff and Bunsen on the coincidences of the dark lines of the solar spectrum with those of certain chemical elements, no foreign astronomer has devoted more consideration to spectrum analysis than M. Secchi of Rome. From his observations, we gather that the spectrum of a Herculis presents to the eye an appearance of a series of fluted columns. The striking contrast between the luminous and the dark portions produces a remarkable stereoscopic effect. The columns are again resolved into finer lines. One line coincides with Fraunhofer's line D of the solar spectrum; two others, one of which is larger and double, with the magnesium line. The spectrum of Persei resembles that of a Herculis. M. Secchi has formed the following conclusions as the result of his investigations so far. 1. The type of the white, or bluish stars, like Sirius and a Lyræ, is marked by a broad band near the place of the line F, and another at the beginning of the violet end of the spectrum. Occasionally a third band, with very fine lines, is visible in the spectra of the brightest stars of this type near the extreme end of the violet. 2. Spectra with large bands, as in a Orionis, Antares,

&c., and stars generally of a red or orange tint; these are often confounded with the next class. Aldebaran comes between the two classes. 3. The type of our sun, comprising Capella, Arcturus, Pollux, &c. Their spectra is covered with fine lines, and direct measurement has shown that they occupy the place of the principal lines of the solar spectrum. An accessory type is that of the constellation Orion, in which all the stars have very similar spectra, characterised by a greenish tint, and in which the line F is very thin. M. Secchi remarks that "each type prevails in one region of the heavens; the green in Orion, the yellow in Cetus, the blue in the Pleiades, Ursa Major, Corona Borealis, &c., while the distinct type represented by a Lyræ includes nearly half of the stars which I have submitted to spectrum analysis."

The spectrum of y Cassiopeia, as observed by M. Secchi, exhibits an appearance very different from that of any other star of a similar colour. For example, in the great majority of white stars, the line F is very clear and broad, as in the spectra of a Lyræ, Sirius, &c. Instead, therefore, of this ordinary dark line in the spectrum of y Cassiopeia, there is in its place a very fine luminous line which is considerably more brilliant than any other portion of the spectrum. By the aid of a micrometer this bright line was found to coincide exactly with the position of the line F. It is premature to decide whether this peculiar spectrum is only a type of many others, or whether it is an isolated example of a star with a physical constitution different from that of its neighbours. Further observations will probably settle this point when the light of most of the visible stars has been systematically analysed. At present, however, it is certainly unique in its character, excepting that the luminous line makes this spectrum something analogous to that of the remarkable variable star in Corona Borealis, a brief description of which we have given in the next paragraph (814). There is, however, this difference in the spectra of these two stars, that whereas the bright line in that of y Cassiopeia has only been satisfactorily observed by one astronomer, the peculiar spectrum of the variable in Corona Borealis has been seen and the position of the bright lines measured, with respect to the corresponding dark lines of the solar spectrum, by several observers in different countries.

Mr. Huggins has discovered that there is a change in the refrangibility of certain of the absorption lines in the spectra of some of the principal stars when compared with corresponding lines in metallic or gaseous spectra; and he accounts for the change by supposing that it is caused by the approach or recession of the star in the line of sight. This important discovery has opened a new field of astronomical research.

814. Remarkable spectrum of a star in Corona Borealis, This extraordinary object suddenly appeared as a star of the second magnitude, on the 12th of May, 1866, near › Coronæ. Several observers noticed it independently both in Europe and America. Its brightness began to diminish from the day of discovery, decreasing at the average rate of half a magnitude per day. By the second week in June, the star was only of the ninth magnitude. It has been identified with an object observed by M. Argelander, of Bonn, and is catalogued as No. 2765, zone +26°, in his Bonner Sternverzeichniss as of the 9 magnitude.

Mr. Huggins was the first to observe the spectrum of this star. He found it of the most remarkable character, unlike that of any celestial body which he had hitherto examined. When viewed with the spectroscope, its light was found to be compound, emanating from two different sources. "Each light forms its own spectrum. In the instrument these spectra appear superposed. The principal spectrum is analogous to the sun, and is evidently formed by the light of an incandescent solid or liquid photosphere, which has suffered absorption by the vapours of an envelope cooler than itself. The second spectrum consists of a few bright lines, which indicate that the light by which it is formed was emitted by matter in the state of luminous gas.' 11 *

The double spectrum of this star has been observed at different observatories, confirming completely the observations of Mr. Huggins. Many explanations or speculations have been given as to the origin of this apparently sudden outburst; but before any definite opinion can be settled, we must wait for further observations. Possibly the star may really be one of the class of variables, with a regular, but unknown, period of maximum and minimum brightness. In the autumn of 1866, its magnitude had again increased to between the seventh and eighth. From observations made with the Greenwich transit-circle, its mean place in the heavens for January 1, 1866, is:

815. Catalogue of Variable Stars. The following list of variable, or periodic, stars has been compiled from the most recent observations, more especially, however, from the catalogue published by Dr. Schönfeld, of Mannheim.† It will be at once seen that the table is more complete than that contained in paragraph 687. The right ascensions and north polar distances are given for the epoch 1870.

Proceedings of the Royal Society, vol. xv. No. 84.

† Catalog von veränderlichen Sternen, mit Einschluss der neuen Sterne. 1866.