406
HISTOEY OF PHYSICAL ASTBONOMY.
me with having made, by mere accident *. Having directed his atten-tion to a large white wall , at the moment of the total immersion of aneclipse of the sun, he saw the moon’s shadow pass upon the wall, tingedwith different colours.” \
The three observations just cited are the only ones that have beenfound in the records of solar eclipses, confirmatory of the phe-nomenon remarked exclusively in France , during the total eclipse of1849. The question naturally arises, to what physical cause are thesingular undulations noticed on such occasions to be ascribed? It isreasonable enough to suppose that, towards the time of total obscuration,the equilibrium of the atmosphere is very much disturbed by the changeof temperature consequent upon the rapid diminution of the visiblesegment of the solar disk. Hence will arise a sudden intermingling ofstrata of air of different densities, and consequently of different refractivepowers, which may obviously tend to produce a succession of undulatorymovements similar to those actually perceived. It is not improbable alsothat the continuous obstruction of the solar light by innumerable foreignsubstances floating in the atmosphere, and subject to incessant displace-ments from the agitation of the air, may contribute towards producingthe observed effect. The extreme smallness of the solar crescent cannotfail to render the interposition of any such particles of matter, morereadily perceptible than it would otherwise be. It is to be remarked alsothat the effects in all such cases will be magnified by diffraction. Thatthe phenomenon is to a certain extent dependent on the operation of thisprinciple, would seem to be indicated by the variety of colour whichdistinguishes the moveable shadows. Moreover, if light be propagated bymeans of undulations in a highly-elastic medium, it is not improbablethat, as the origin of light rapidly diminishes in magnitude, causes ofinterference may arise leading to a conflict of undulatory movements,whence may ultimately result phenomena analogous to those alluded toin the foregoing observations.
A phenomenon which seems to be of a totally different nature from anyof those hitherto alluded to, has been occasionally observed during solareclipses, especially such as are of the annular species. In the latter case,previous to the formation of the annulus, the western limb of the moonexhibits an indented appearance, resembling a succession of heads.Almost immediately the beads become elongated, assuming the aspect oflong, black, parallel streaks, uniting the limbs of the sun and moon. Inthe next instant these lines give way, as if they had been snappedasunder by the eastward motion of the moon, and the annulus thenappears completely formed. A similar phenomenon ensues in a reverseorder, when the eastern limb of the moon is approaching the sun’s limb,
* “ Par le plus grand hasard du monde.”
f Memoires pour servir a VHistoire et au Progres de I’Astronomie, p. 258, 4to., St.Petersbourg, 1738. M. Arago states that having searched, page by page, the volumes ofthe various academic collections, he found only one observation corroborative of thosemade in France relative to the undulations which occurred on the occasion of the totaleclipse of 1842; but that he was unable to cite it in detail, in consequence of the looseleaf, to which he had consigned the memorandum of it, having been mislaid.(Annuaire , 1846, p. 399.) It is not probable that either of the two Swedish observa-tions cited in the text was that which fell under his eye; since, if he had met with theone, the other could hardly have escaped him. It is evident, at any rate, that theobservation cited by Delisle, in a work which does not form part of any academiccollection , cannot be the one referred to by the illustrious philosopher.