290 PROCEEDINGS OF THE AMERICAN ACADEMY 



by a column of air extending to the upper regions of the at- 

 mosphere, the temperature of which he supposed to have been 

 raised, according to the principles of Espy's theory, forty de- 

 grees Fahrenheit, by the condensation of the aqueous vapor 

 previously contained in it ; and he inferred the existence, in 

 this case, of a force capable of elevating bodies of considera- 

 ble weight. 



He stated that he had not been able to test the electrical 

 theories by a similar process of computation, for want of suf- 

 ficiently definite ideas of their nature. He thought, however, 

 that indications of electrical action in tornadoes were so strong, 

 as to make it very desirable that electricians should investigate 

 the data requisite for such a computation. 



Dr. W. F. Channing was satisfied that the causes usually 

 assigned were insufficient to account for the mechanical effects 

 of tornadoes, and was glad that a calculation of the forces of 

 various theories had been undertaken by Professor Peirce. 

 " The rotation in these storms is insufficient to produce a 

 vacuum in the axis of the whirl, adequate to the elevation of 

 heavy bodies by the means of the in-rushing and ascending 

 air. In a water-spout which he had observed on Narraganset 

 Bay, in 1845, the rotation of the trunk was obviously too slow 

 to account for the elevation of the water which took place in 

 the axis. An observer at Somerville had distinctly seen the 

 rotation of the column of the tornado at West Cambridge, a 

 mile and a half or two miles distant. When the smallness of 

 the diameter of this revolving column (a few hundred feet) 

 was considered, the velocity of rotation was at once reduced 

 to a rate insufficient to account for the mechanical effects 

 produced. 



" In the beautiful experiment exhibited by Professor Guyot 

 at the last meeting of the Academy, there was a permanent 

 cause of rotation in the upper strata of the revolving fluid ; 

 that fluid was inelastic, and it was confined within the walls 

 of a cylinder, which prevented the dispersive effect of the 

 centrifugal force. In the case of the tornado, there is no 



