months. From October, A.D. 626, to the following June, a defalcation of light to the extent of one-half is recorded; and in A.D. 1547, during three days, the sun is said to have been so darkened that stars were seen in the day-time. Now of all these instances, supposing them all to have been owing to spots, either unusually large or numerous, there are only two, those of A.D. 807 and 1607, which deviate so much as two years from the epochs of maximum fixed as above.

(394 b.) Sir W. Herschel (Ph. Tr. 1801), considering the appearance of abundant spots on the sun's disc as evidence of an agitated state of its gaseous envelope, and regarding the extrication of light and heat as the results of chemical processes likely to be promoted by the more intimate mixture of heterogeneous materials having mutual affinities, has attempted to show, though from very imperfect records (such as alone could be procured by him at that date) that years of remarkably abundant or deficient spots have been also remarkable respectively for their high or low general temperature, and especially for abundant and deficient harvests. The point has been inquired into by M. Gautier*, who from an assemblage of meteorological averages obtained in thirtythree stations in Europe, and twenty-nine in America during eleven years of observation, finds a trifling preponderance (0°11 Fahr.) in the opposite direction. On the other hand. M. Wolf, in the memoir above cited, from an examination of the Chronicles of Zurich from the year A.D. 1000 to A.D. 1800, is led to a conclusion in accordance with this speculation, and considers them as affording decisive evidence" that years rich in solar spots are in general drier and more fruitful than those of an opposite character, while the latter are wetter and more stormy than the former."

(394 c.) Although more properly belonging to the domain of general physics than of astronomy, it is impossible to omit mentioning here the singular coincidence of this period of the recurrence of the solar spots with that of those great disturbances in the magnetic system of the earth to which the epithet of" magnetic storms" has been affixed. These

Bibl. Univ. de Geneve. 1844.

disturbances, during which the magnetic needle is greatly and universally agitated (not in a particular limited locality, but at one and the same instant of time over whole continents, or even over the whole earth), are found, so far as observation has hitherto extended, to maintain a parallel both in respect of their frequency of occurrence and intensity in successive years with the abundance and magnitude of the spots in the same years, too close to be regarded as fortuitThe coincidence of the epochs of maxima and minima in the two series of phenomena amounts indeed to identity, a fact evidently of most important significance, but which neither astronomical nor magnetic science is yet sufficiently advanced to interpret.

ous.

(395.) Above the luminous surface of the sun, and the region in which the spots reside, there are strong indications of the existence of a gaseous atmosphere having a somewhat imperfect transparency. When the whole disc of the sun is seen at once through a telescope magnifying moderately enough to allow it, and with a darkening glass such as to suffer it to be contemplated with perfect comfort, it is very evident that the borders of the disc are much less luminous than the center. That this is no illusion is shown by projecting the sun's image undarkened and moderately magnified, so as to occupy a circle two or three inches in diameter, on a sheet of white paper, taking care to have it well in focus, when the same appearance will be observed.* This can only arise from the circumferential rays having undergone the absorptive action of a much greater thickness of some imperfectly transparent envelope (due to greater obliquity of their passage through it) than the central.-But a still more convincing and indeed decisive evidence is offered by the phæno

* This has been denied by Arago on the evidence of certain phænomena observed with his "polariscope;" but the fact is so palpable, that it is matter of some astonishment that it could ever fail to strike the most superficial observer. The matter has been placed beyond a doubt, however, by direct experiments both photometric and thermic. The details of the latter by Sig. Secchi will be found in Astron. Nashr. Nos. 806. 833, and go to prove that the calorific radiation of the center of the sun's disc is nearly double of that from its borders, and that the equatorial regions are somewhat hotter than the polar (Comptes Ren dus. Aug. 26th, 1852).-(Note added 1858.)

mena attending a total eclipse of the sun. Such eclipses (as will be shown hereafter) are produced by the interposition of the dark body of the moon between the earth and sun, the moon being large enough to cover and surpass, by a very small breadth, the whole disc of the sun. Now when this takes place, were there no vaporous atmosphere capable of reflecting any light about the sun, the sky ought to appear totally dark, since (as will hereafter abundantly appear) there is not the smallest reason for believing the moon to have any atmosphere capable of doing so. So far, however, is this from being the case, that a bright ring or corona of light is seen, fading gradually away, as represented in Pl. I. fig. 3., which (in cases where the moon is not centrally superposed on the sun) is observed to be concentric with the latter, not the former body. This corona was beautifully seen in the eclipse of July 7. 1842, and with this most remarkable addition-witnessed by every spectator in Pavia, Milan, Vienna, and elsewhere: three distinct and very conspicuous rose-coloured protuberances (as represented in the figure cited) were seen to project beyond the dark limb of the moon, likened by some to flames, by others to mountains, but which their enormous magnitude (for to have been seen at all by the naked eye their height must have exceeded 40,000 miles), and their faint degree of illumination, clearly prove to have been cloudy masses of the most excessive tenuity, and which doubtless owed their support, and probably their existence, to such an atmosphere as we are now speaking of. In the total eclipse of July 28. 1851, similar rose-coloured protuberances were observed, one in particular of a form quite decisive as to their cloudy nature, rising straight up verti

sents the appearance as seen by Prof. Schmidt, at Rastenburg, just as a column of smoke rising in calm air is often seen to be drifted off horizontally when it has attained such a height as to bring it into an upper current of wind. To complete the resemblance, a detached and perfectly insulated mass B of the same rosy colour was observed at some distance from the drifted train A, which was connected with another mass C, by a narrow red band or streak D.

(395 a.) The existence of such an atmosphere superior to the luminous envelope being admitted, affords an easy explanation of the faculæ, considered as vast waves in the photosphere (art. 388.). In an atmosphere consisting of strata gradually decreasing in density, any cause of undulation acting on the inferior strata will throw them up to a vastly greater height, and therefore produce far greater waves in them than would arise from the same cause acting on the surface of a definite ocean of liquid matter, by reason of their being partially sustained against gravity, leaving their inertia free to carry them up to a higher level. The experiment is easily tried in oil floating on water, or in saline solutions increasing in density downwards, and is at once amusing and instructive. [See Note (O) § (395 b b.)]

(396.) That the temperature at the visible surface of the sun cannot be otherwise than very elevated, much more so than any artificial heat produced in our furnaces, or by chemical or galvanic processes, we have indications of several distinct kinds: 1st, From the law of decrease of radiant heat and light, which, being inversely as the squares of the distances, it follows, that the heat received on a given area exposed at the distance of the earth, and on an equal area at the visible surface of the sun, must be in the proportion of the area of the sky occupied by the sun's apparent disc to the whole hemisphere, or as 1 to about 92,000. A far less intensity of solar radiation, collected in the focus of a burning glass, suffices to dissipate gold and platina in vapour. 2dly, From the facility with which the calorific rays of the sun traverse glass, a property which is found to belong to the heat of artificial

S

fires in the direct proportion of their intensity.

3dly, From the fact, that the most vivid flames disappear, and the most intensely ignited solids appear only as black spots on the disc of the sun when held between it and the eye.† From the last remark it follows, that the body of the sun, however dark it may appear when seen through its spots, may, nevertheless, be in a state of most intense ignition. It does not, however, follow of necessity that it must be so. The contrary is at least physically possible. A perfectly reflective canopy would effectually defend it from the radiation of the luminous regions above its atmosphere, and no heat would be conducted downwards through a gaseous medium increasing rapidly in density. That the penumbral clouds are highly reflective, the fact of their visibility in such a situation can leave no doubt.

(397.) As the magnitude of the sun has been measured, and (as we shall hereafter see) its weight, or quantity of ponderable matter, ascertained, so also attempts have been made, and not wholly without success, from the heat actually communicated by its rays to given surfaces of material bodies exposed to their vertical action on the earth's surface, to estimate the total expenditure of heat by that luminary in a given time. The result of such experiments has been thus announced. Supposing a cylinder of ice 45 miles in diameter, to be continually darted into the sun with the velocity of light, and that the water produced by its fusion were continually carried off, the heat now given off constantly by radiation would then be wholly expended in its liquefaction, on the one hand, so as to leave no radiant surplus; while on the other, the actual temperature at its surface would undergo no diminution.‡

By direct measurement with the actinometer, I find that out of 1000 calorific solar rays, 816 penetrate a sheet of plate glass 0-12 inch thick; and that of 1000 rays which have passed through one such plate, 859 are capable of passing through another. H. 1827.

†The ball of ignited quicklime, in Lieutenant Drummond's oxy-hydrogen lamp, gives the nearest imitation of the solar splendour which has yet been produced. The appearance of this against the sun was, however, as described in an imperfect trial at which I was present. The experiment ought to be repeated under favourable circumstances. Note to the ed. of 1833. According to the more recent experiments of Messrs. Fizeau and Foucault, the intensity of the light at the surface of Drummond's lime-ball is only one-146th part of that at the surface of the sun!—(Note added 1858.)

"Results of Astronomical Observations at the Cape of Good Hope,” p. 444.