654 
Proceedings of the Royal Society 
On all the evenings on which this first glow was distinct, I ob- 
served a thin film of silvery cloud, if cloud it could be called, form 
or become visible over the western sky, just after the sun had dis- 
appeared, It was rather curious that this filmy cloud seemed to 
have a definite boundary underneath, its wavings or undulations 
being quite distinct. The peculiar silvery appearance of this filmy 
cloud struck me at the time, as it had a strange lustre about it ; but 
it was not till I considered the necessity for crystalline reflectors fully 
to explain the peculiarities in the visibility of the first glow, that it 
struck me these crystals would also explain the peculiar silvery 
lustre of this haze or cloud. This crystalline dust would also 
account for the great brilliancy of the mid-day glare, accompanied 
as it was with comparatively little white light at a distance from 
the sun, and with fair transparency in the other parts of the sky, 
the crystals between us and the sun reflecting more light than those 
situated in other directions. 
So far as these observations have gone, no relation has been 
detected between the brilliancy of these after glows and the 
humidity of the atmosphere, as given either by the spectroscope, or 
by the wet and dry bifib thermometers. The observations, so far 
as tliey go, rather indicate a dry atmosphere as favourable to 
brilliancy, but the observations are too few to settle the point. 
We shall now suppose that our first glow has sunk down and 
melted into the horizon colours. The, sky now has little red in it 
anywhere save in the glowing west. The sky is now very much 
the same as before the first glow began, only the light from all 
parts is now much less. With the assistance of these glowing 
colours in the west, we shall now try to explain the way in which 
the second glow is produced. Shortly after the first glow dis- 
appeared, it was observed that the overhead changed in appearance, 
the blue slowly faded and became a whitish (X)lour. In a short 
time the white changed, to a redd^h hue, and at last the whole 
heavens glowed with a fine red light. Though this second glow 
was only now visible, it had been there, for some time, hut could 
not he seen by our unassisted eyes. |f, however, we used the 
polariscope we could see it at any time, after the brilliancy of the 
first glow had gone. It was not visible to the unassisted eye, for 
the same reason that the first glow could not he seen overhead, on 
