518 



1 believe that these results are even more accurate than 

 those furnished by the anemometer ; and they prove that in 

 this locality the direction of the wind was exactly reversed, 

 and, therefore, that the centre of the vortex passed over the 

 College. 



" A remarkable circumstance connected with the direction 

 of the fallen trees is their great uniformity, the individual di- 

 rections seldom differing more than 10° from the mean. This 

 is an indirect evidence of the great violence of the gale ; and 

 it proves, moreover, that the transition from the south-east to 

 the north-west wind was immediate. There is greater regu- 

 larity in the direction of the trees fallen from the north-west 

 than in those which have been blown down from the opposite 

 quarter. This may have arisen partly from the greater vio- 

 lence of the gale in the former direction ; but it is partly also 

 due to the circumstance that the trees which fell from the 

 north-west are generally larger and in a less inclosed portion 

 of the ground. It may be mentioned also, that the trees which 

 fell from the north-west generally lie to the southward of the 

 others ; there are, however, two large trees in the garden 

 lying side by side, but in directions diametrically opposed. 



" It has been stated that in the College Park the shifting 

 of the wind amounted to 180° ; and it has been inferred that 

 the centre of the vortex passed over that spot. From what 

 has been said as to the nature of the phenomenon, it will fol- 

 low that in other localities, over which the vortex did not pass 

 centrally, the wind must have shifted through diflferent points 

 of the compass, and through angles smaller in proportion to 

 their distances from the centre. Thus, on the southern side 

 of the line described by the centre of the vortex, the change 

 of the wind should be from south to west, and on the northern 

 side of the same line from east to 'north. We are not yet in 

 possession of facts which bear upon this point ; but from the 

 limited dimensions of the vortex, and the consequent smallness 

 of the distance necessary to produce such a variation, it is 



