2 
PHYSICS: E. THOMSON 
1. Streamers seen in auroras, singly or in composite masses, are in 
reality vertical, or approximately so, to the earth's surface, nearly 
parallel when adjacent, and only slightly divergent even when miles 
apart; the divergence being due to curvature of the earth's surface. 
2. In any aurora, the streamers appear to be located in bands or 
zones more or less wide in latitude extending generally in east and west 
direction, or forming belts or zones between parallels of latitude in 
which the streamers extend vertically upward like trees in a forest. 
3. In some rare auroras the vertical streamers are closely limited to 
a narrow belt of latitude; sometimes only 2 or 3° or even less, in width 
north and south, while the east and west extent of the narrow belt 
may be very great. 
4. In wide spread auroras the belt of vertical streamers may cover 
great extents of latitude and extend east and west unlimited distances. 
This appears to have been the case in the recent great aurora of August 
26, 1916. 
5. The curvature of the so called auroral arch is an optical effect of 
perspective, slightly added to by the curvature of the earth. The 
appearance of folded curtains of streamers is a similar effect of super- 
position and perspective when the active band or zone covers many 
degrees in latitudes. It is probable that the lower ends of auroral 
streamers have about the same height in the earth's atmosphere; a 
layer from which they stream upward to heights which vary in different 
displays or even in the same display. This layer probably exists at a 
height of about fifty miles and conducts laterally or horizontally, thus 
distributing the electricity discharged from it into the streamers. 
6. The convergence of long streamers toward the zenith seen in the 
greater auroras, is purely an optical effect of perspective, the streamers 
being vertical. 
7. The so called zenith crown is in reality due to bundles of streamers, 
nearly vertical but seen on end. They appear as patches of shifting 
light in or near the zenith, sometimes surrounded by apparently con- 
verging streamers from the north, east, and west, and even from the 
south; converging in appearance only. 
8. The convergence of streamers is of the same nature as the con- 
vergence of straight parallel railway tracks in the distance, or better, 
the apparent convergence toward the sun, of the sunbeams seen in dust- 
laden air, when the sun itself is obscured by a small irregular cloud, or 
is back of a broken mass of clouds. 
In April, 1883, at near the middle of sunspot period and coincidently 
with the occurrence of an enormous sunspot, there was perhaps the 
