HISTORY OF PHYSICAL ASTRONOMY.

403

by any known physical principles. That it is connected in some way withthe magnitude and relative position of the segment of the sun which hasnot undergone immersion, would seem obviously to follow from its modeof variation; hut it is difficult to assign any cause why the moon’s surfaceshould appear more distinct on the side on which the sun is visible, thanon the opposite side.

A phenomenon of a somewhat similar nature, but one admitting of asatisfactory explanation, was noticed by M. Arago during the total eclipseof 1842. Forty minutes after the commencement of the eclipse, while aconsiderable portion of the sun was still visible, he observed the dark con-tour of the moon projected upon the bright sky; it corresponded exactlywith the portion of the limb that appeared upon the sun; lie remarkedalso, that it was brighter at the parts where it met the solar disk*.He ascribes the phenomenon to the projection of the moon upon the solaratmosphere, the brightness of which, by an effect of contrast, renderedthe outline of the moon’s dark limb discernible. This explanation ac-counted for the fact of the limb being brighter towards the sun, since itappeared from all the observations of total eclipses, that the luminousring exhibited the greatest concentration of light at those parts whereit bordered upon the solar disk.

The contour of the dark limb of the moon was also visible to M. EugeneBouvard at Digne, and to M. Flaguerges at Toulon ; but it is remarkablethat no other observer of the total eclipse of 1842 makes mention of asimilar appearance.

Halley remarked, that during the total eclipse of 1715, he observed per-petual flashes, or coruscations of light, which seemed for a moment to dartout from behind the moon, now here and now there, on all sides, butmore especially on the western side a little before the emersion f. Louvillealludes to a similar phenomenon in his account of the same eclipse; hestates, that during the time of the total obscuration, it seemed as if therehad been a succession of trains of gunpowder constantly exploding on the

so favourable for observation, that if the whole of the solar disk had undergone immer-sion, the dark and bright parts of the moon’s surface would have been as plainly dis-tinguishable as Ferrer states they were on the occasion of the total eclipse of 1806.According to M. Arago, the light by which the bright parts are visible in such a case,must exceed the light of the more obscure parts, by at least a sixtieth part. Now,at the commencement of the eclipse, the moon’s surface is illuminated unequally by thelumiere cendree, or ashy light, and equally by the atmospheric light diffused by the fullsun; and the former of these lights (or rather the excess of the light emitted by thebright parts over the light of the more obscure parts) bears so small a proportion to thelatter, that the whole of the moon’s surface is almost equally illuminated, and the variousirregularities by which it is diversified will be completely effaced. As the eclipse, how-ever, increases in magnitude, the light diffused by the visible segment of the solar diskdiminishes at an equal rate, and in consequence, the light of the lumiere cendree con-tinually bears an increasing ratio to the atmospheric light. At length, the latter is somuch enfeebled, that the excess of the light of the bright parts of the lunar disk over thelight of the more obscure parts, arising from the lumiere cendree , amounts to the sixtiethpart of the atmospheric light, whereupon, according to the above remark, the irregularitiesof the lunar disk will become distinguishable. The American observer states, that themoon’s surface became distinctly visible as soon as one-eighth of the solar diameter wasimmersed. It is difficult to conceive that at so early a stage of the eclipse the atmosphericlight would be so much enfeebled, that the lumiere cendree would form a sensible frac-tion of it. Admitting, however, that the phenomenon was due to such a cause, itfollows that the lunar surface will appear more illuminated towards the sun, on accountof the greater intensity of the atmospheric light, except at the part bordering on the solarsegment, where it will naturally appear dark from the effect of contrast.

* Annuaire, 1846, p. 372. f Phil. Trans., 1715, p. 249.

D D 2