80 On the Height of the Aurora Borealis. 



parts of the earth,where, at times, the au', during severe frosts 

 is rendered so dry and non-conducting, that tho electricity is 

 prevented from escaping from the clouds and vapour to the 

 earth in those parts ; and therefore accumulates, till it flashes 

 back through the upper and rarer air towards the warmer 

 and moi'e negative parts of the earth, and thus causes the 

 aurora. 



The opinion so generally received, that the aurora is at 

 times at such heights as to be far above the limits of our at- 

 mosphere, is so directly opposed to this theory, that one or 

 the other must be erroneous ; I therefore suggested that the 

 observations which give such altitudes to the aurora, may 

 be incorrect, either from some error in the observations, or 

 from some other luminous meteor being taken for an aurora, 

 or from some optical illusion in the observations. 



At the conclusion of my paper, Professor Challis stated 

 that he had (at Cambridge), in conjunction with Professor 

 C'hevallier (at Durham), taken the altitude of an aurora, whicli 

 gave positive proof of its being at least 160 miles high, and 

 consequently far above the limits of the atmosphere. I, of 

 course, could not doubt the cori*ectness of these observations ; 

 but as I was not convinced that my own views were wrong, I 

 have been led by several reasons to conclude that the eri'or 

 must be owing to some optical illusion. 



In the first place, the fact, that the auroral corona is never 

 seen except in or near the zenith, shews that it is an appear- 

 ance only, and not in reality a convergence of auroral streams 

 to one point, as in that case it would at times be seen at every 

 point of the compass, according to the situation of the obser- 

 ver ; and auroral streamers would at times be seen crossino- 

 the sky in an oblique direction, instead of always rising nearly 

 perpendicular to the horizon. 



It is shewn by Lieutenant Hood, that auroral arches are 

 only so in appearance, " as the aurorte whicli filled the sky at 

 Cumberland House, from the northern horizon to the zenith, 

 with wreathe^ and flashes, assumed the shape of arches at 

 some distance to the southward ;" and I believe the reports 

 on the aurora of the 24th of October, by Professors Challis 



