256 ASTRONOMY. 



in such sort, that those near the zenith appeared more distant from each 

 other than those near the horizon, a certain indication that the real 

 distances of the rows were either partly or exactly the same. And it 

 was farther observable, that during the rapid lateral motion of the beams, 

 their direction in every two nearest rows was alternate, so that whilst 

 the motion of the one row was from east to west, that of the next was 

 from west to east. 



The cause of the northern lights is unknown. It is supposed, and 

 there are some facts which apparently support the hypothesis, that they 

 are to be attributed to the passage of electricity throughout the higher 

 regions of the atmosphere. At present, however, the subject is involved 

 in much obscurity ; andf rom the very few opportunities that philosophers 

 have of seeing the aurora, it is extremely difficult to obtain accurate data 

 to reason upon. 



CAUSES OF WHIRLWINDS, MOVING PILLARS OF SAND, 

 AND WATERSPOUTS. 



To the principle of electricity we refer a class of remarkable phenomena 

 not usually ranked together whirlwinds, pillars of sand, and waterspouts 

 the appearance of which being very well known from frequent 

 description, we shall at once proceed to consider their nature. In the 

 explanations generally given it is assumed that there are currents of air 

 blowing in different directions, the oblique meeting of which causes an 

 eddy or vortex, having a vacuum in its interior. Against this hypothesis 

 it may be objected that, in the greater number of instances recorded, the 

 air has been either calm or with a wind moderate and steady without any 

 cross currents. If these meteors had a mechanical origin of this kind, 

 they ought to abound most where variable winds and storms prevail, as 

 on sea-coasts, near headlands, and among hills. On the contrary, they 

 are most rare in such cases, rather affecting climates and seasons of hot 

 still atmosphere, in desert plains or tropical seas. Besides, in order to 

 form a vortex, it is necessary that a coherent body be present to deflect 

 the current into the tangential motion producing the whirl. A vortex 

 cannot be formed in the free atmosphere, whatever be the respective 

 velocities or angle of meeting of currents, and, according to all experience, 

 a shift of wind is preceded by a calm, lasting until one of the currents 

 has obtained predominance. That waterspouts and whirlwinds are 

 independent of motion in the air, is made evident by their having often 

 a rapid progression although the air around them be still, and by their 



