80 PHYSICAL GEOGKAPHY OF THE SEA, AND ITS METEOROLOGY. 



greater rapidity, or else a part of it must be sloiighed off above, 

 and so turn back before reaching the cahns about the poles. The 

 latter is probably the case. Such was the conjecture. Subsequent 

 investigations* have established its correctness, and in this way : 

 they show that the south-east trade-winds, as in the Atlantic, blow, 

 on the average, duiing the year, 124 days between the parallels 

 of 25^ and 30^ S., and that as you approach the equator their 

 average annual duration increases until you reach 5^ S. Here 

 between 5^ and 10^ S. they blow on the average for 329 out of the 

 365 days. 



218. Now the question may be asked, Where do the supplies 

 pUeVaie'^suuihS't which furnish air for these winds for 329 days come 

 trade-wind in the from ? Thc " trades " could not convey this fresh 

 cross the band 253. supply of air across the parallel of 25"^ S. diuring 

 the time annually allotted for them to blow in that latitude. 

 They cannot for these reasons : (1.) Because the trade-winds in 

 lat. 5^ are stronger than they are in lat. 25°, and therefore, in 

 equal times, they waft larger volumes of air across 5° than they 

 do across 25°. (2.) Because the girdle of the earth near the equa- 

 tor is larger than it is farther off, as at 25° ; therefore, admitting 

 equal heights and velocities for the wind at the two parallels, it 

 would, in equal times, bear most air across the one of larger cir- 

 cumference. Much less, therefore, can the air which crosses the 

 parallel of 25^ S. annually in the 124 trade-wind days of that 

 latitude be sufficient to supply the trade-winds with air for their 

 329 days in lat. 5^. Whence comes the extra supply for them in 

 5° ? (3.) Of all parts of the ocean the trade-winds obtain their 

 best development between 5° and 10° S. in the Atlantic Ocean, for 

 it is there only that they attain the unequalled annual average 

 duration of 329 days. But referring now to the average annual 

 duration of the south-east trade-wind in aU seas, we may, for the 

 sake of illustration, liken this belt of winds which encircles the 

 earth, say between the parallels of 5° and 25^ S., to the frustum of 

 a hollow cone, with its base towards the equator. 



219. Now, dividing the winds into only two classes, as winds 

 Winds with north- with %OTtliing and winds with southing in them, 

 southnVin'the'm^ actual obscrvatious show, taking the world around, 

 contrasted. ^-^^^^^ wiuds liaviug southiug in them blow into the 

 southern or smaller end of this cone for 209 days annually, and 



* Nautical Monographs, No. 1, Observatory, Washington, October, 185D. 



