TORNADO IN SPRUCE CREEK VALLEY, PENNSYLVANIA. 459 



that intense electrical excitement generally accompanies tornadoes ; bnt, 

 while it is easy to see how this may be the effect of a commotion of the 

 atmospliere, it is very difficult to understand, on the known principles 

 of electiicity, how it can be the cause of such violent ])lienomena. 

 Electricity generally exists in nature in a state of equilibrium, and the 

 discharges which we witness are due to the restoration of the equili- 

 brium, while, on the other hand, as it api)ears to me, all the jdienomena 

 which are exhibited hud a ready explanation on well-known thermal and 

 pneumatic principles. Let us first consider the condition of the atmos- 

 phere previous to the coming on of the tornado. The air was close 

 and sultry; that is, it was surcharged with moisture, which, absorbing 

 the rays of the sun, rendered it unusually warm and abnornuilly light. 

 If, in this condition, we suppose a stratum of colder wind from the north- 

 v.est, the direction from which the meteor moved, to be passing above, 

 wo shall have a condition of atmosi)here possessing the potential energy 

 requisite to produce the phenomena observed. As the upper wind passed 

 over the earth at a considerable elevatior, the natural equilibrium would 

 be disturbed, a heavier stratum being above, a lighter one below. The 

 equilibrium would be of an unstable character, and the sliglstest irreg- 

 ularity at a given si)ot would induce the rushing up of the air at the 

 point of least resistance, and a descent around this point of the heavier 

 stratum. The column of agitation would be more circumscribed if a 

 whirling motion were given the mass, and the whole would be carried 

 forward by tlie motion of the n[)per current. The moist air would rush 

 in below from all sides, and, ascending in the vortex and mingling with the 

 colder stratum above, would instantly be converted into visible vapor. 

 If the moist stratum had been sufliciently thick and the upward DU)tion 

 sufficiently violent to carry the vapor above the snow-line of the lati- 

 tude of the place, the drops of water would have been frozen, anxl 

 probably thrown out on each side of the vortex, giving rise to two 

 tracks of hail. According to this hy[)othesis the electricity is due to 

 the condensation of the vapor, or, more detinitely, to the formation of a 

 vertical water-conductor, of which the natural electricity is disturbed 

 by the induction of the plus electricity of s[)ace, and the minus elec- 

 tricity of that of the earth below. The great mechanical effects which 

 are exhibited in tornadoes are readily accounted for on the principle of 

 continued pressure or a succession of i'mpulses, as an illustration of 

 which we may mention the effect produced by blowing on a light ball 

 in a hollow tube. In this case the ball is followed by a (;ontinued pres- 

 sure from one end of the tube to the other ; at every moment it receives 

 a new impulse, which it retains by its own inertia, and finally leaves 

 the tube with the accumulated effect of the Ibrce which is ap})lied to it 

 through its whole course. In like manner, a stratum of air set in motion 

 by the removal of pressure in front of it, while a pressure is continued 

 in the rear, is impelled forward with an accumulating velocity, and 

 finally acquires an energy sufficient to overcome obstacles of astonishing 



