125 



which they are subjected, in consequence of the multiplicity of points 

 which their twigs and leaves furnish for the emission of the electrical 

 fluid. The fact that the leaves of trees thus injured, appear after- 

 wards as if they had beeen partially scorched, seems to countenance 

 this idea. The twisting of the chimney at New Brunswick, as above 

 mentioned, seems difficult to explain, agreeably to the idea of a gene- 

 ral whirl throughout the' whole area of the tornado track. The 

 chances are infinitely against any chimney having its axis to coincide 

 with that of a great whirlwind, forming a tornado; and it must be 

 evident, that in any other position, it could only be subjected to the 

 rotary force on one side at a time. But if this were adequate to twist 

 the upper upon the residual portion, the former would necessarily be 

 overthrown. Evidently, it could not be left, as was the chimney 

 which called forth these remarks. 



During the tornado at New Haven, chimneys seemed to be espe- 

 cially affected. One, after being lifted, was allowed to fall upon a 

 portion of the roof of the house to which it belonged, at a distance from 

 its previous situation too great to have been reached, had it been 

 merely overthrown. In the case of a church which was demolished, 

 a portion of the chimney was carried to a distance greater than it 

 could have reached without being lifted by a vertical force. 



It appeared quite consistent that chimneys should be particularly 

 assailed, since that rarefaction, which, by operating upon the roofs of 

 houses, carries them away, must previously cause a great rush of 

 air through the chimney flues. But this concentration of the air must 

 tend to facilitate the " convective"* discharge in that direction ; since 

 an electrical discharge by a blast of air, is always promoted by any 

 mechanical peculiarities favouring an aerial current, or jet. 



That during a recent tornado in France, articles were carried from 

 the inside of a locked chamber to a distance without, when no opening 

 existed besides that afforded by a chimney, seemed to justify the sug- 

 gestion that there must be a great rush of air through such openings.! 



* A "convective" discharge, or a discharge by "convection," in the very 

 appropriate language of the celebrated Faraday, is a process by which electri- 

 city is conveyed by the transfer of electrified bodies from one excited surface 

 to another in an opposite state. This is conceived to be a good definition of 

 the discharge which produces a tornado. 



t Dr. Hare did not conceive it proper to trespass upon the time of the So- 

 ciety, to make any allusion to that part of his Memoir, in which the three 

 enormous concentric spaces occupied by the earth, the denser non-conducting 

 atmosphere, and the rare conducting medium beyond the denser atmosphere, 



