103 



to magnetism, occurred in the early history of the development of 

 the law of storms, and has not yet, so far as I am aware, been dis- 

 tinctly refuted by the public, or withdrawn by its promulgator. 



In Colonel Reid's first work (1838) on the revolving motion of 

 the hurricanes, after having, in the earlier portion, detailed, in the 

 most satisfactory manner, the laws of the 'phenomena, he gives, in 

 the latter portion, a glimpse of a theory of them, or at least, details 

 an experiment in which, on the surface of a magnetised iron shell 

 representing the earth, a rotation in opposite directions was pro- 

 duced in helices in either hemisphere of the ball. This was thought 

 very interesting, as the hurricanes are found to revolve in opposite 

 directions in either half of the world ; and it was further stated 

 that in St Helena, where the magnetic intensity is small, hurricanes 

 are unknown ; while in the West Indies, where hurricanes are so 

 rife, the magnetic intensity is at a maximum. 



Here, it will be observed, is no attempt to shew whether the 

 magnetic power is sufficient to cause the observed effect, or has any 

 power in that way at all, nor even to trace whether this particular 

 coincidence at two points, in the tropical belt of the earth, prevailed 

 at all others also ; and in the Colonel's last publication (1848) the 

 question and the experiment are withdrawn altogether. 



When, however, we examine the subject more extensively, we find a 

 pretty general rule to prevail all round the world, viz., that hurricanes 

 are most frequent in the western parts of those seas where the trade- 

 wind is suddenly stopped by the occurrence of land, and is unknown 

 in the eastern parts of the seas where it begins. Thus, not only is 

 the placid climate of St Helena fully accounted for by being in the 

 eastern position of the South Atlantic, but equally the similar freedom 

 from revolving storms of the Cape De Verde Islands, the NW. and 

 SW. coast of Africa, with California and Peru on the eastern shores 

 of the Pacific. 



And again, while the West Indies are pointed out as likely places 

 for hurricanes, so are Hio Janeiro, Canton, the Mauritius, and 

 Madras, and, in fact, almost every place where hurricanes have 

 been met with. ' ^ 



The stoppage, then, and interference of the trade-wind, a purely 

 mechanical question, is the cause of the hurricanes, and, according 

 to the greater or less force of the trade-wind, and the greater quan- 

 tity of air struggling to get over the barrier, as observed in the case 

 of water when a river is in a flood, or on a sea-coast at spring-tide, 



