Feb. 14, 1859.] . "AURORA BORE ALIS" IN GREENLAND. 110 



witli the magnetic poles. These "bands almost invariably stretch 

 east and west, or a point or two on each side, never going beyond 

 north-west and south-east or south-west and north-east. This is 

 the case whether the wind be north, south, east, or west. "With 

 north and south winds the bands still hold their east and west 

 direction, but are curved to leeward, like any long, yielding body 

 would be if secured at both ends. 



With regard to height, extent, and density of the aurora, I 

 have observed as follows: — 1st, as to height, about 10 miles west 

 from my place of observation is a mountain 4000 ft. high ; I have 

 never seen the aurora or its rays descend below this, and consider 

 it never occurs nearer the earth than between 4000 and 5000 feet, 

 though its altitude is extremely varying, ranging probably from 

 that distance to 10 or more miles in height. The most certain 

 means of estimating the height I found was to observe the effect of 

 the aurora on clouds, and vice versa. I have seen it with the cumuli. 

 When the aurora is above these, the clouds appear as dark masses 

 on it, and hide the light ; but when the aurora is under the clouds 

 (that is, nearer the earth than the clouds), the light is reflected by 

 the clouds, and passing through the aurora makes the latter appear 

 more bright in that part ; the whole under surface of the cloud is 

 also completely illuminated. 



The splendid red and green colours of the aurora are in general 

 only seen when the air is clear and free from vapour and clouds ; 

 these colours are not quite so intense as the solar iris, but con- 

 siderably brighter than the lunar. 



Eegarding the speed of " auroral " motion (by which I mean 

 the transmission of a wave in the band of light from one part of the 

 band to another, independent of wind), I have remarked on several 

 occasions, when the aurora was partly above and partly below the 

 clouds (which I estimated 10,000 ft. high), that a wave moved from 

 a part of the band, crossing the summit of a mountain to the zenith, 

 in periods of from 4 to 8 seconds, rarely half a minute. The moun- 

 tain was 4000 ft. high, distant 10 miles, height of eye of observer 

 24 ft. Supposing the height of the clouds to be judged with mode- 

 rate accuracy, we could calculate approximately the speed of the 

 ** auroral " motion. The wind has no influence on this motion, 

 since the waves of light run equally rapid and frequent to windward 

 or leeward. 



By reference to the journal it will be seen when the aurora was 

 above or below the clouds : the temperature, barometer, and weather 

 being given, will enable any one interested in this subject to judge 

 of the nature of the clouds, and estimate nearly their height. 



