Mr. W. Brown on the Storms of Tropical Latitudes. 209 



the height of the barometer would not be applicable; the se- 

 cond is evidently unequal to the effect ; the last therefore is- 

 the only one in which a whirlwind, such as that assumed, can 

 be supposed to be maintained. Now if the air moved in a perfect 

 whirl, it is evident that no alteration of pressure in any part 

 of the atmosphere within it would take place, the air would 

 revolve at a constant distance from the centre; but this is not 

 the case in the present instance; the air is not brought back 

 on its revolution to exactly the place it previously occupied, 

 but to one a little more remote from the axis; because it is 

 found that storms enlarge in diameter as they advance : now 

 the only rarefaction which the centrifugal force can effect, is 

 that occasioned by the air thus increasing its distance from 

 the axis, and therefore it is limited by the amount of the in- 

 crease of the diameter of the storm. Now let the whole path 

 of a hurricane be 3000 miles, and let its diameter on attain- 

 ing the end of it be doubled, then a rate of progress of eighteen 

 miles an hour will give only an increase of T £ H th of the dia- 

 meter in one hour; this enlargement therefore produces the 

 whole amount of the diminution of the air within the whirl- 

 wind (whose height is not supposed to extend very far above 

 the earth), which is required to be restored by the flow of the 

 air of the higher strata of the atmosphere into the rarefied 

 portion, an amount so small that this flow must be almost 

 adequate to its entire restoration, so that only a decrease in 

 the pressure of the air, almost trifling compared with what 

 actually occurs, could take place from this cause. 



But it is evident that the only direct proof of a rotative ac- 

 tion must result from the accordance between the directions 

 of the wind required by the hypothesis and those found by 

 observation. But all the observations so laboriously collected 

 by J. P. Espy, go to show that the direction of the wind is 

 totally irreconcileable with the existence of a gyratory motion ; 

 and although they may be viewed with some suspicion on ac- 

 count of their being advanced in defence of a favourite theory, 

 his results are too consistent with each other to be set aside. 

 But independently of these, the facts given in Colonel Reid's 

 volume afford results, of a different kind from those obtained 

 by Espy, equally opposed to this theory. 



If the direction of the wind were taken at an equal number 

 of stations upon radii of a whirlwind equally distant from 

 each other, the numbers representing the directions of the 

 wind for the several points of the compass, making allowance 

 for the progressive motion, would be equal. Now it is ob- 

 viously impossible that this test could be rigorously applied 

 to any hurricane; but when a great number of observations 



Phil. Mag. S. 3. Vol. 23. No. 151. Sept. 1843. P 



