/ 
G5-6 Proceedings of the Royal Society 
particles near the horizion ; it is therefore necessary for me to show 
how this second glow is produced without the direct rays of the sun. 
It may be that while the second glow is visible, the sun is still 
shining in the atmosphere overhead^ but at the time of the 
second glow it touches only the upper limits, and there are no 
particles there large enough to reflect red light. Indeed, the second 
glow is not visible till the rays have passed out of the air in which 
the particles are large enough to reflect blue light. It is clear, 
.therefore, that the second glow cannot have its source in direct 
illumination. 
There are different ways of satisfying ourselves that the second 
glow is but a reflection of the sunset colours on the horizon, by the 
same particles as shone by direct sunlight in the first glow. Before 
going further, there is a most important fact which requires 
attention. It is one to which we are so much accustomed that we 
might not give it its true value. It is, however, so important, in the 
study of these phenomena, that it is necessary we should constantly 
keep, it in view. Of course, every one knows that daylight is far 
brighter than gaslight, but how difficult it is to realise the difference. 
On any ordinary day at sundown, light the gas ; it has no effect — 
the. room is not a bit better lighted. Leave the gas liL and as the 
sun sinks, note how the gas, begins to light up a wider and wider 
area, and at last the room appears to be brilliantly illuminated by 
the gas alone, while outside we can atill see our way about, and the 
last of these sunset effects are still visible. Now, try to realise this 
enormous scale of brilliancy we. have got to deal with in daylight. 
We should be more sensible of the difference were it not for the 
curtain in front of our eyes, which nature draws closer and closer the 
more brilliant the light is. As only enough, light is admitted to the 
retina to give distinct vision, and as the amount is regulated by 
unconscious movements, we are not so. sensible as we. might be of 
the vast scale of illumination used by nature. Keeping ever in view 
this vast scale of brilliancy we have to work with at sunset, it is 
easy to see that what is dark at one time, and under certain con- 
ditions, may really appear brilliantly illuminated a short time after 
under different conditions. A cloud, for instance, on a bright sky 
may look black, but remove the bright sky and. we find the cloud is 
brilliantly lighted up. 
