660 Proceedings of the Poycd Society 
the red glow comes to ns mixed with a certain amount of blue light 
from the sky beyond, a certain amount of red light is masked, and 
it is only when the red is greater than a certain proportion that it 
becomes visible as red. Any amount less, only changes the blue 
towards white. If, however, on searching the sky overhead after 
sunset with the polariscope, we find the red still there, we may 
be sure it will be follovt^ed by a red glow in the west, though it 
may not be visible to our eyes. 
Conclusions. 
We may sum up our conclusions under the five following 
heads : — 
1. If our atmosphere were perfectly pure, and free from suspended 
particles, there Avould be no twilight. When the sun sunk under 
the horizon, its rays would pass through our atmosphere into space, 
without revealing themselves by illuminating our sky, and the 
moment the sun disappeared total darkness would follow, almost 
as quickly as when a candle is extinguished at midnight. 
2. The greater the amount of suspended particles — up to a certain 
quantity — ^there is in the atmosphere, and the greater the height to 
which they are diffused, the more the western sky will be illuminated 
at sunset, and the longer will be the twilight. 
3. The greater the amount of suspended particles of extremely 
small size there is in the atmosphere, the more highly coloured will 
be the transmitted light of the sun and the reflected light of the 
sky. 
4. The western afterglow is caused by the transmitted light 
being reflected to the earth by small dust crystals floating in the 
air. 
5. When there are plenty of suspended particles in the atmo- 
sphere, the western sky at sunset becomes so brilliantly luminous 
that it radiates light in every direction, and illuminates the suspended 
particles all over the sky, causing them to shine long after the 
direct rays of the sun have left them. 
