274 Rotary Action of the Providence Tornado. 



here greatly reduced, and gives fewer prostrations than under 

 either the first or third quadrants. The minimum of effect oc- 

 curs on the arrival of the Une e k, at the rear of the second quad- 

 rant. 



But on the arrival of the third quadrant, the prostrating power 

 on the left side becomes more and more efficient by the ceasing 

 of the backward and the accession of the progressive movement ; 

 and at or near the hne of /m, it again takes effect, with rapid 

 increase. The destructive force is also much augmented here 

 by the greater velocity of the heart of the whirl, near its axis, 

 and the impetus must rapidly increase in energy to its maximum 

 effect, as at w w o, taking off any tree which may here remain, 

 and carrying aloft, or sweeping onward, the objects previously 

 prostrated on the line ex k. 



If a tree on the left side of the track falls on one previously 

 thrown down by the tornado, the last fallen will also have the 

 more onward direction, as on the other side : unless both have 

 fallen in the second quadrant, where few prostrations occur. — The 

 fourth quadra7it, for causes noticed in considering the ^rs^, can 

 have little prostrating effect, except perhaps on the small area 

 near its axis. 



If we now conceive of our figure as applied only to the limits 

 of prostration or destruction which constitute the visible path of 

 the tornado, it becomes apparently and relatively unequal, in its 

 right and left hand quadrants, the axis appearing greatly eccen- 

 tric, and in the same degree, at least, as the left band or belt of 

 prostrations is found narrower than that on the right of the axis. 

 This apparent, but illusive form of the whirl, may be illustrated 

 by fig. V ; which is drawn on the same lines with the preceding 

 figure. 



It will readily be seen that this eccentricity of the axis, on the 

 visible track, will be in proportion to the progressive velocity of 

 the tornado ; other things being equal. Thus, if Mr. Allen be 

 nearly right in his estimate of the rate of progress in the Provi- 

 dence tornado, the eccentricity shown in its path would be gen- 

 erally less than is shown in figures IV and Y. On the other 

 hand, if the progressive velocity should be as great as Professor 

 Loomis informs me he ascribes to the tornado of February last, in 

 Ohio, viz. about forty miles an hour, the eccentricity would in 

 such a case be greatly increased, showing the axis as far out- 



