The various parts of our satellite are not all equally reflecting.  Here, it may depend on the form, elsewhere, on the nature of the materials.  Those persons who have examined the moon with telescopes, know how very considerable the difference arising from these two causes may be, how much brighter one point of the moon sometimes is than those around it.  Now, it is quite evident that the relations of intensity between the faint parts and the brilliant parts must continue to exist, whatever be the origin of the illuminating light.  In the portion of the lunar globe that is illuminated by the sun, there are, everybody knows, some points, the brightness of which is extraordinary compared to those around them; those same points, when they are seen in that portion of the moon that is only lighted by the earth, or in the ash-coloured part, will still predominate over the neighbouring regions by their comparative intensity.  Thus we may explain the observations of the Slough astronomer, without recurring to volcanoes.  Whilst the great observer was studying in the non-illuminated portion of the moon, the supposed volcano of the 20th of April, 1787, his nine-foot telescope showed him in truth, by the aid of the secondary rays proceeding from the earth, even the darkest spots.

Herschel did not recur to the discussion of the supposed actually burning lunar volcanoes, until 1791.  In the volume of the Philosophical Transactions for 1792, he relates that, in directing a twenty-foot telescope, magnifying 360 times, to the entirely eclipsed moon on the 22d of October, 1790, there were visible, over the whole face of the satellite, about a hundred and fifty very luminous red points.  The author declares that he will observe the greatest reserve relative to the similarity of all these points, their great brightness, and their remarkable colour.

Yet is not red the usual colour of the moon when eclipsed, and when it has not entirely disappeared?  Could the solar rays reaching our satellite by the effect of refraction, and after an absorption experienced in the lowest strata of the terrestrial atmosphere, receive another tint?  Are there not in the moon, when freely illuminated, and opposite to the sun, from one to two hundred little points, remarkable by the brightness of their light?  Would it be possible for those little points not to be also distinguishable in the moon, when it receives only the portion of solar light which is refracted and coloured by our atmosphere?

Herschel was more successful in his remarks on the absence of a lunar atmosphere.  During the solar eclipse of the 5th September, 1793, the illustrious astronomer particularly directed his attention to the shape of the acute horn resulting from the intersection of the limbs of the moon and of the sun.  He deduced from his observation that if towards the point of the horn there had been a deviation of only one second, occasioned by the refraction of the solar light in the lunar atmosphere, it would not have escaped him.