84 Notices of Tornadoes, Sfc. 



does not take place, founds thus an objection to electrical agency. 

 I conceive that it would be unreasonable to expect a magnetic 

 needle to be affected by an electrified blast of air if protected from 

 its mechanical force. 



It has been shown by Faraday that without peculiar manage- 

 ment, tending to prolong the reaction, the most delicately sus- 

 pended needle cannot be made to diverge in obedience to the 

 most powerful discharges of mechanical electricity. An electri- 

 cal spark may impart a feeble magnetism, but is too rapid and 

 transient, to afiect a needle. Moreover, when a needle is at right 

 angles to an electric current which would be quite competent to 

 influence it if parallel to it, there can be no consequent move- 

 ment, since the current tends to keep it in that relative position. 

 The direction of every electrical discharge inducing a tornado 

 must necessarily be nearly at right angles to the needle, since it 

 must be vertical, while the needle is necessarily horizontal when 

 so supported as to traverse with facility. 



I do not perceive any facts or suggestions in the article by 

 (Ersted, which are competent to render the phenomenon of which 

 he treats more intelligible than it was rendered by the accurate sur- 

 vey and examination of the track of the New Brunswick tornado, 

 by Pres. A. D. Bache, and Mr. Espy, in connexion with the 

 accounts published by other witnesses of that and other similar 

 meteors. 



It seems to be admitted on all sides, that within a certain space 

 there is a rarefaction of air tending to burst or unroof houses ; 

 that the upward blast consequent to this rarefaction, carries up all 

 movable bodies to a greater or less elevation ; and that an afflux of 

 air ensues from all quarters to supply the vacuity which the ver- 

 tical current has a tendency to produce. Trees within the rare- 

 fied area are uprooted and sometimes carried aloft, but on either 

 side of it, or in front, or in the rear, are prostrated in a direction 

 almost always bearing towards a point which durmg some part 

 of the time in which the meteor has endured has been under the 

 axis of the column which it formed. 



It appears to me that all the well authenticated characteristics 

 enumerated by (Ersted, are referable to the view of the case thus 

 presented. This distinguished author assumes that there is a 

 whirling motion, although between American observers, this is a 

 debated question. It seems in the highest degree probable that 



