Rotary Action of the Providence Tornado. 275 



ward, perhaps, as would be in line with m, ox: f I, in these two 

 figures. 



Fig. V. 



From this examination it appears to result, that an observer 

 who follows the track of a tornado after its departure, will find 

 on one side of the apparent axis of its path, if it be a whirlwind, 

 a continued series of prostrations pointing almost invariably on- 

 ward and inward, with various degrees of inclination to the 

 course of the path. While on the other side of the axis, a nar- 

 rower band or belt of prostrations will be found, which are also 

 inclined mainly inward and onward, but showing greater incli- 

 nations from the line of progress, together with frequent cases 

 which incline more or less backward and sometimes even out- 

 ward from the course of the tornado. 



It may also appear, that a want of proper attention to the ne- 

 cessary conditions of the prostrating power in a progressive whirl- 

 wind, can alone induce us to ascribe such effects to supposed an- 

 tagonistic winds, blowing simultaneously in opposing directions. 



Leaving, for a moment, the more tangible features of this in- 

 quiry, we may now take some notice of the more outward por- 

 tions of the "cone" or whirlwind, which are supposed not to be 

 comprised in figure IV. Assuming here the involuted and in- 

 ward motion, with its upward discharge at the centre, it follows 

 that the impulsive accession of air which is necessary for main- 



