OCEANIC CIRCULATION. 259 



of the winds prevalent in certain regions ; and when I 

 first wrote on the Grulf Stream there was no evidence on 

 the subject even approaching Maury's (or that collected 

 by Maury) in accuracy and completeness. But there is 

 one argument which those who have adopted the trade 

 winds as the primary cause of the Grulf Stream appear 

 to me to have overlooked, and it is on this argument 

 that my own view has been chiefly based. The trade 

 wind zone of the northern hemisphere is not constant 

 in position; but travels northwards and southwards 

 with the northerly and southerly motion of the sun in 

 declination. The change in the position of the zone of 

 calms is not, indeed, so great as is stated in Buchan's 

 meteorology, where it is said to travel from 25 north to 

 25 south of the equator ; but it is considerably greater 

 than was supposed by Dove, Kaemtz, and others. If we 

 set the extreme shift of the northern trade-zone at ten 

 degrees we are certainly not over-rating it. Taking 

 this zone as extending in spring or autumn from 10 to 

 25 north latitude, we should have it in winter extending 

 from 5 to 20, and in summer from 15 to 30, the 

 only part common to these two ranges being that from 

 15 to 20 that is to say, the northern five degrees 

 of the winter zone, and the southern five degrees of the 

 summer zone, each zone being 15 wide. Now, if any 

 one will mark these zones on the North Atlantic, he 

 will find that while the zone of winter trades would 

 produce a current flowing into the southern half of the 

 Gulf of Mexico, the zone of summer trades would pro- 

 duce a current flowing into the northern half. The 



8 2 



