of Edinhurgh, Session 1883-84. 
659 
light, and shone more brightly than the true glow ; and on some 
evenings, these thin clouds and the glow haze seemed to be so near 
the same elevation, that it was difficult to distinguish the one from 
the other. 
It is scarcely necessary to say that the clouds, like the second 
glow, did not really get lit up with red light. The appearance of 
lighting up was not a reality, but was due to the other light dying 
out of the sky more quickly than the red light in the west, and the 
light now coming from the clouds being more brilliant than that 
from the sky, the clouds appeared bright on the dark sky. Some 
interesting colour effects were observed on the second illumination 
of the clouds. They were not always coloured to the same depth 
of red at all points where they were exposed to the western glow. 
The western edge of the cloud, for instance, was generally of a light 
red; whilst the parts which projected underneath, and were exposed 
to the western glow, were of a much deeper red. This was caused 
by the western edge of the cloud being exposed to the illumination 
of the white and blue sky in the west and overhead ; this light 
with the red made the exposed edge whitish red, while the parts 
projecting underneath the clouds were protected from the white and 
blue, and exposed only to the horizon glow ; they shone therefore 
with a redder light. 
It has been said that the colour of the first glow was not the same 
as the sunset colours of the horizon, but was of a crimson hue. 
Now this is just what we might expect, because, from the lower 
strata in which the large particles are, red or reddish light is reflected 
to us, but this red light is accompanied by a certain amount of 
blue light from the higher strata, and the combination of a smalt 
amount of blue with the red gives a crimson tone. The second 
glow, for the same reason, was of a similar colour. On the horizon 
the red is combined with the green and yellow, causing it to appear 
more or less orange. 
It will be interesting in the future, when all these peculiar 
sunsets are over, to see if we can find any of this afterglow by 
means of a polariscope. It is obvious that there may be a con- 
siderable amount of red afterglow in the west, without its ever 
being visible to us. The reason for this, as already explained, 
is, that a mixture of red and blue lights appear white, and as 
