HISTORY OF PHYSICAL ASTRONOMY.

373

taking into account the modifying influence of extraneous causes, selectingfor this purpose the region ■which is vertical relative to the spectator.He remarks that every particle of the atmosphere, although illuminatedmainly by the direct rays of the sun, is also affected in some degree bythe light reflected in every direction from the other particles. Duringthe progress of the eclipse, the region of the atmosphere which lies verti-cally above the spectator, ceases gradually to be illuminated by the directrays of the sun, while, on the other hand, it is constantly exposed to thesame intensity of reflected light from the region which lies in the horizonof particles situate at a great altitude above the place of observation, forthe sun still continues to shine upon that region with all his force. Theproportion of direct to reflected light continues to diminish as the eclipseincreases in magnitude, until at length the reflected light produces amore intense effect than the direct, and thereby determines the visiblecolour of the sky. Now it is a fact which cannot fail to have come underthe observation of every person, that the rays of light proceeding fromthose regions of the atmosphere that lie near the horizon, invariably differin hue from those which are transmitted from the more elevated regions.It follows, therefore, as a necessary consequence, that the colour of thesky in the region which is vertical relative to the spectator will undergoa perceptible change*.

This explanation is very ingenious, but it does not give a satisfactoryaccount of the unnatural aspect which the sky exhibits during an eclipse,and which has been universally remarked to be totally different from theappearance of the dawn or twilight. It is to be borne in mind, however,that the circumstances which determine the faint illumination of theatmosphere during an eclipse, although analogous to those upon whichthe phenomenon of the dawn or twilight depends, are not absolutelyidentical with them; and hence it is not improbable that the differentcondition of the light, arising from this cause, may produce the pallid huewhich is visible during an eclipse.

It has been found that while the colour of the sky changes very sen-sibly during eclipses of the sun, a similar effect is also produced uponterrestrial objects. Even as early as the year 840, a.d., it was remarkedthat during the total eclipse of the sun which happened in that year, thecolours of objects on the earth were changed f. Kepler mentions thatduring the solar eclipse which happened in the autumn of 1500, thereapers in Styria noticed that everything had a yellow tinge J. It wasremarked by MM. Plantade and Clapies, on the occasion of the totaleclipse of 1706, that as the obscuration increased or diminished, objectschanged their colour. When two-thirds of the solar diameter wereeclipsed, they assumed the colour of orange yellow ; when there was onlyabout the twenty-fifth part of the diameter visible they assumed a reddishtinge, resembling water that has been diluted with wine.

Sir John Clarke, in an account of the annular eclipse of 1737, states,that there was no considerable darkness, but that the ground was coveredwith a kind of dark greenish colour §.

In Le Gentil ’s brief description of the total eclipse of 1766, it is statedthat during the greatest obscuration objects assumed a tinge of lividyellow, which produced a very remarkable effect JJ.

* Annuaire, 1846, p. 296. + Ad Vitellionera Paralipomena, p. 2 r ‘i.'

t Ad Vitellionem Paralipomena, p. 303. § Phil. Trans., 1737, 1 1 197.

|| Voyage dans les Mers de l’lnde, tome ii., p. 16.