649 • 
of Edinburgh, Session 1883-84. 
second made its appearance. This second upper glow was not, 
however, confined to the west, but the whole heavens gloWed with a 
strange reddish light, of very much the same tone of colour as the 
first upper glow. This second one generally seemed to attain its 
maximum first in the east, then all over the sky^ a short time after 
which it died awayi, and the sky took up its usual night appearance. 
Only a very few observations of this Second glow were obtained, as 
it was only on a few evenings that its appearance and dis- 
appearance were sufficiently distinct for the hour to be correctly 
noted. The time after sunset, when it appeared and disappeared', 
cannot therefore be given with any accuracy. I may state, how- 
ever, that it appeared within a few minutes after the first upper 
glow had vanished, perhaps about 30 or 35 minutes after sunsetj 
and it did not last long. It is obvious that the times here given 
of the first and second glows only refer to the particular cases 
observed. A month or two earlier, when there was more dust in 
the air, there is every reason for supposing that both the first and 
second glows would have continued much longer after sunset, as 
there would then be plenty of dust in the higher regions to sift the 
rays, also to refiect the red ones, and the sun would thus continue 
to shine on the dust to a later hour. It is necessary for me to 
state that the position from which the observations were made was 
such that the sun did not set on the sea, but over land, and that 
the land was low, and the position of observation high enough 
for the sun to be seen down to the true horizon. 
The conclusions at which I have arrived are — 1st, that the first 
upper glow is produced by the direct rays of the sun illuminating 
the particles of matter suspended in the atmosphere j and 2nd, that 
the brilliant western horizon light is the source of illumination of 
the second upper glow^ 
I shall now attempt to state the manner in which all the different 
successive colour effects are produced. The sun’s rays, in passing 
through our atmosphere, there encounter innumerable multitudes of 
very small dust particles. Some of these particles are so ex- 
tremely small that they can stop and scatter only the rays of small 
wave length- that is, those of the violet end of the spectrum. 
The result of this is, that the blue light is stopped and reflected in 
every direction, -while the rays of the red end of the spectrum are 
