connects the observer with the sun; and, (3,) that the variations of color, altitude, &c., according to the relative positions of the sun and moon, with respect to the observer, may be accounted for by the fact that, owing to the interference of the solar rays diffracted in passing the mountains on the moon's edge, some of the ether particles from which light is sent to the observer, are more strongly excited than those adjacent.

Quite recently the generally received theory of the physical constitution of the sun has been called in question by the celebrated Leverrier. The observations of this astronomer, at Tarazona, in Spain, during the total eclipse of July 18th, 1860, indicate, in his view, the necessity of abandoning the old hypothesis and adopting one more simple. He doubts the existence of the various solar envelopes proposed by Herschel and others; thinks the light of the sun due simply to its high . temperature; and regards the solar clouds, or flame-like protuberances, as a stratum of roseate matter diffused over its surface. This new and bold hypothesis will doubtless be subjected to a searching and rigorous criticism. A great number of observations will probably be required before a perfectly satisfactory explanation of the phenomena can be given. When, therefore, we take into consideration the fact that total eclipses of the sun are of rare occurrence and short duration, we cannot reasonably expect a speedy solution of the mystery. Mr. Nasmyth, of England, has suggested that these protuberances might possibly be rendered visible by an artificial total eclipse of the sun. This plan, however, has not been suc

cessful.

Granting that the corona is the atmosphere of the sun, its hight from the surface of the photosphere cannot be less than half a million of miles. If now we suppose, what is at least highly probable, that the heat of our central luminary was once much greater than at present, the extent of his atmosphere must also have been greater in a corresponding degree. Admitting it to have extended to the point at which the centrifugal force would balance the central attraction,* and to

The present distance of this point from the center of the sun is about sixteen millions of miles.

have gradually contracted from the loss of heat, the equatorial portions of the gaseous spheroid would be abandoned, and would revolve separately round the sun. The appearance of such a zone or ring of atmospheric molecules would probably be similar to that of the zodiacal light-the well known nebulous aurora frequently observed either shortly before sunrise, or after sunset, and which was ascribed by Cassini to the blended light of an innumerable multitude of extremely minute asteroids revolving round the center of our system. It is proper to add, however, that some astronomers now regard the zodiacal light as produced by a terrestrial ring.

From comparisons of the horizontal and vertical measures of the sun's diameter, Lalande, Lindenau, and Delambre concluded that the polar diameter was greater than the equatorial, or that the figure of the sun was that of a prolate spheroid. According to Lindenau the difference of the two diameters was nearly one-four hundredth of the greater. Recently, however, Professor Winlock has thoroughly discussed the original observations of Bradley, Maskelyne, and Piazzi,-from which this improbable conclusion had been derived,—and has found the equatorial diameter rather greater than the polar; the ellipticity being one-three thousandths, or about one-eleventh that of the earth. This result is confirmed, moreover, by the observations of Struve, Bessel, and Hubbard, as well as by the later Greenwich and Edinburgh observations. It may be remarked that even this slight degree of ellipticity is greater than that indicated by theory.

Besides the sun's revolution on his axis, and a small motion. round the center of gravity of the solar system, he is now known to have a rapid progressive motion in space. As lately as the commencement of the eighteenth century the stars called fired were believed to maintain absolutely the same relative positions. Dr. Halley first discovered, in 1717, by comparing the observed right ascensions and declinations of Sirius, Aldebaran, and Arcturus, with their positions as given by Ptolemy, that they had undergone a very sensible displacement. This discovery was soon confirmed by other observers; the proper motions of a considerable number of stars being detected by a comparison of modern observations alone.

About the middle of the last century, (1748), the celebrated Bradley suggested that the sun himself might have a translatory motion in absolute space, and that the apparent changes of position among the other stars might be thus partially accounted for. Lalande inferred the probability of the sun's progressive motion from the fact of its rotation; since the latter, without the former, could not be communicated by a single mechanical impulse. Sir William Herschel, who devoted much attention to this subject, adopted the hypothesis of the translatory motion of the solar system. From an elaborate discussion of the changes which had been detected in the relative positions of the stars, he arrived at the conclusion that the sun, with all the planets and comets of our system, was moving, in 1790, towards a point whose right ascension is two hundred and sixty degrees and thirty-four minutes, and north declination twenty-six degrees and seventeen minutes. From the nature of the case, however, this determination was only approximate.* More recently the question has been ably investigated by Argelander, Otto Struve, Gauss, and Galloway. The close agreement between the results of these independent researches is decisive, both as to the fact of the sun's motion and its present direction. Its period, however, as well as the form and dimensions of the mighty orbit which it describes, remains to be determined. These interesting problems are now demanding the earnest attention of astronomers; and, although their inherent difficulties may long baffle investigation, we cannot reasonably doubt their ultimate solution.

Various hypotheses have been advanced in regard to the origin of solar heat. Professor Thomson, of England, supposes it to be produced by the fall of meteoric matter. Insuperable objections, he maintains, lie against all other theories hitherto proposed. The fall of meteors to the earth's surface establishes the fact that immense numbers of these bodies are constantly traversing the planetary spaces. It has been calculated that not less than ten millions enter the earth's atmosphere every day. Reasons are not wanting for supposing

* This point is in the constellation Hercules. Herschel's Memoir was communicated to the Royal Society in 1783.

their numbers to increase with great rapidity as we approach the center of the system. But on account of the greater force of gravity at the sun's surface, the heat produced by the fall of a given amount of meteoric matter will be much greater than at the surface of the earth. It can hardly be doubted, therefore, that at least a portion of the sun's heat is generated by the fall of meteors into his atmosphere; but whether the theory is sufficient to account for all cannot now, perhaps, be satisfactorily determined. As bearing upon this subject, the following facts, stated by the President of the British Association for the Advancement of Science, at the meeting of 1860, are exceedingly interesting:

"On the first of September last, at 11h. 18m. A. M., a distinguished astronomer, Mr. Carrington, had directed his telescope to the sun, and was engaged in observing his spots, when suddenly two intensely luminous bodies burst into view, on its surface. They moved side by side through a space of about thirty-five thousand miles, first increasing in brightness, then fading away. In five minutes they had vanished. Momentary as this remarkable phenomenon was, it was fortunately witnessed and confirmed, as to one of the bright lights, by another observer, Mr. Hodgson, at Highgate, who, by a happy coincidence, had also his telescope directed to the great luminary at the same instant.

"It may be, therefore, that these two gentlemen have actually witnessed the process of feeding the sun, by the fall of meteoric matter; but however this may be, it is a remarkable circumstance, that the observations at Kew show that on the very day, and at the very hour and minute of this unexpected and curious phenomenon, a moderate but marked magnetic disturbance took place; and a storm, or great disturbance of the magnetic elements, occurred four hours after midnight, extending to the southern hemisphere. Thus is exhibited a seeming connection between magnetic phenomena and certain actions taking place on the sun's disk—a connection, which the observations of Schwabe, compared with the magnetical records of our Colonial Observatories, had already rendered nearly certain."

Within the last thirty-five years the observations of Schwabe,

Sabine, Wolff, and Lamont have established the fact of a mysterious connection between the solar spots and terrestrial magnetism. From the epoch of any minimum in the number of the sun's maculæ, a nearly regular increase is observed for about five and a half years, when, a maximum being reached, the number and magnitude again diminish during an equal period; the entire cycle from minimum to minimum, or from maximum to maximum, occupying, according to Wolff, eleven years and one-ninth. A corresponding cycle has also been found in the magnetic variations. These periods have the same length. The maxima and minima of the one coincide with those of the other, and a like agreement is observed throughout both series of changes. Such coincidences cannot be ascribed to chance. As yet, however, the nature of the connection can only be conjectured.*

M. Buys-Ballot, of Utrecht, has found, from an elaborate discussion of a great number of meteorological observations, that there is a short period of variation in the amount of solar heat received by the earth; the period from maximum to maximum being 27.682 days. The time of the sun's rotation, with respect to the earth, is 27-2 days. Hence the variations cannot be ascribed to the solar spots, or to any inequality in the heating power of different portions of the sun's surface. Buys-Ballot has suggested that it may be produced by a ring of nebulous or meteoric matter revolving round the sun in a period slightly exceeding that of the sun's rotation, and therefore immediately exterior to the circle of equilibrium between the sun's attraction and the centrifugal force due to his rotation. Such a nebulous zone, but little inclined to the plane of the earth's orbit, would probably influence our temperature by absorbing or retaining a portion of the solar heat.

* These results in regard to the magnetic variations, have been confirmed by the investigations of A. D. BACHE, LL. D., and CHARLES A. SCHOTT, Esq. See the "Discussion of the Magnetic and Meteorological Observations made at the Girard College Observatory, Philadelphia, in 1840, 1841, 1842, 1843, 1844, and 1845."-Smithsonian Contributions to Knowledge, Vol. XI.

sun.

The radius of this circle is about one-sixth of the earth's distance from the