LET. XXII.

SUN AND MOON. 393

observe exactly either the beginning or end ofa lunar eclipse, even with a good telescope;for the earth’s shadow is so faint and ill-definedabout the edges, that when the moon is eitherjust touching or leaving it, the obscuration ofher limb is scarcely sensible. But both the be-ginning and end of solar eclipses are visible in-stantaneously ; for the moment the edge of themoon’s disc touches the fun’s, his roundnessseems a little broken on that part; and themoment she leaves it, he appears perfectly roundagain.

The moon, when totajly eclipsed, is seldominvisible, but generally appears of a dusky co-lour, resembling tarnished copper, which somehave thought to be owing to her own nativelight; but the true cause of this appearance isthe scattered beams of the fun, which are sobent into the earth’s shadow, in their passagethrough the atmosphere, as to afford us a suffi-cient quantity of light to render the moon visi-ble. There have, however, been eclipses of themoon, when in that part of her orbit nearest theearth, in which she has entirely disappeared; butthese instances are very rare. Hevelius men-tions one of this kind, which happened on the25th of April, 164.2, when he was not able todistinguish the place of the moon, even with agood telescope, although the sky was sufficientlyclear for him to see stars of the fifth magnitude.

From