28 On the 'prevailing Storms of the Atlantic Coast. 



3. The regular and obvious proportion which the several diame- 

 ters of the storm and whirlwind, and their rate of progression, bear 

 to their duration, at each point over which they pass. 



4. The different and opposite, or nearly opposite direcUons in 

 which the wind is found to blow upon the opposite sides of the track, 

 and also upon the opposite marginal portions,, of both the storm and 

 the whirlwind. 



The last consideration, if established, hardly falls short of demon- 

 strative evidence of the supposed identity in the mode of acnon in 

 these different masses of moving atmosphere. Every person, on ex- 

 amining the track of a destructive whirlwind, where it has passed 

 through a forest, will, in crossing that track, often find the trees pros- 

 trated in exactly opposite directions, and it is obvious that this effect 

 must necessarily follow, as the result of the acknowledged cause, a 

 circular, or rotative force in the whirlwind. The same effect was 

 equally apparent, only on a larger line of observation, after the storm 

 or hurricane of 1821, as already described. The same general evi- 

 dence of a sudden or a progressive change in the direction of the 

 wind, runs through all the accounts which we have given, or which it 

 is in our power to submit, in relc^tion to other storms. 



In relation to whirlwinds of the smaller class, we may here take 

 occasion to remark, that it is not conceived to be essential to the 

 character of a whirlwind, that its axis of rotation should occupy a ver- 

 tical position, or one but slightly inclined to the plane of the horizon.- 

 On the contrary, the axis, or center of gj^ation, in whirlwinds of a 

 limited character, may, and probably often does, occupy a horizontal 

 position at a considerable height in the atmosphere. This variety of 

 whirlwind is presumed to enter largely into the formation of thunder- 

 storms and squalls, and particularly hail-storms. 



Having attempted to establish the circumrotative character of the 

 south-east storm which has been described, we are led to inquire 

 whether other south-easterly storms possess the like character ; and 

 whether this be also an attribute of the north-eastern storms of our 

 coast, and also what ccnsiitutes the specific difference of character 

 in these storms. 



If the foregoing views be sufficiently established, it must follow, 

 that the direction of the ivind at a particular place, forms no part of 

 the essential character of a storm., hut is only incidental to that par- 

 ticular portion or parallel, of the rout or track of the storm ivhich 

 may chance to become the point of observation. We have seen that 



