426 Dr. Hare on the Theory of Storms, 



takes place without a centripetal force resulting from an 

 hiatus. 



20. But the author has not informed us how these unequal 

 or opposing forces are generated in the atmosphere. Without 

 any assigned cause, he appeals to " certain opposite and un- 

 equal forces by which a rotative movement of unmeasured vio- 

 lence is produced." This rotative movement, although alleged, 

 as above, to be an effect, is stated immediately afterwards to 

 be " the only known cause of violent and destructive winds or 

 tempests." 



21 . In a memoir on the causes of tornadoes and water-spouts, 

 and in some subsequent communications published in the 

 Transactions of the American Philosophical Society, and re- 

 published in Silliman's Journal, various facts and arguments 

 were mentioned, tending to prove that the proximate cause of 

 the phaenomena of a tornado is an ascending current of air, 

 and the afflux of wind from all points of the compass to sup- 

 ply the deficiency thus created. 



22. In this mode of viewing the phaenomena, no difference 

 of opinion exists between Bache and Espy and myself, how- 

 ever we may differ respecting the cause of the diminution of 

 atmospheric pressure within the track of a tornado, which 

 gives rise to the ascending current. 



23. I adduced several facts, upon the authority of the skilful 

 survey made by those gentlemen, proving that the effects were 

 in some cases inconsistent with the existence of a whirl; and I 

 mentioned one which could not be explained without attribu- 

 ting it to a gyratory force. I was led to consider gyration as a 

 contingent, not an essential feature in the meteors in question. 



24-. It appeared reasonable to suppose that the connection 

 of confluent streams of air rushing towards an axis moving 

 progressively, might be productive of a whirling motion. 

 The contortion of six feet of the upper part of a brick chim- 

 ney upon the lower portion, so as to cause the corners of either 

 portion to project over the sides of the other, was deemed in- 

 explicable without ascribing it to gyratory force. Subse- 

 quently, however, it occurred to me that this fact was more 

 likely to be the result of a local than of a general whirl; since 

 in the latter case, the chimney could not have been twisted as 

 described, without being precisely at the centre of the whirl- 

 wind. That such could have been its position, appears to 

 me extremely improbable; and had it been so situated, as the 

 whirlwind was estimated to be moving progressively at the 

 rate of seventeen miles per hour, it is to me incomprehensible 

 how the portion which was dislocated could have escaped an 

 overthrow. Evidently, although twisted upon its base while 



