38 Professor Mitchell on tJie Trade-Winds. 



(a) The definiteness of the boundary by which the trade- 

 winds are limited and separated from currents flowing in an 

 opposite direction, and that they commence at once in full vi- 

 gour at that boundary, are circumstances of great weight. " Thus 

 in the northern Atlantic, from the same limits whence the north- 

 east trade blows towards the equator, a south-west (or rather 

 west south-west) wind not uncommonly prevails in the contrary 

 direction. So in the southern Atlantic, from the limits of the 

 south-east trade, the prevalent winds are nearly converse, (west- 

 north-west). Now adverting to these winds blowing contrariwise 

 from the same limit, there is difficulty to conceive the origin 

 of either trade, but as derived from upper strata of the atmo- 

 sphere, and if that source of supply at the commencement be ac- 

 knowledged, there is little reason for rejecting it in the wind's 

 subsequent progress." This author, however, attributes the trade- 

 winds to the diminution of the air's specific gravity by absorp- 

 tion of moisture.* 



(Jb) In the Sandwich Islands, the trade-wind blows from the 

 north-east. Upon the summit of Mouna Kea, -f* in Hawaii, esti- 

 mated to be more than 18,000 feet in height, Mr Goodrich, 

 in the month of April, found a wind from the south-west resem- 

 bling the cold blustering winds of March, in New England J. 

 On the Peak of Teneriffe, Humboldt, Von Buch, and others, 

 have encountered a raging west wind which scarcely allowed 

 Humboldt to keep his feet. This was in summer. In the win- 

 ter this west wind descends to the coast §. These facts show 

 that the currents of the upper atmosphere are strictly counter 

 currents, which carry eastward the air, the trade-winds have car- 

 ried westward. They do not seem to be a mere result of motion 



• Colebrooke's Meteorological observations in a voyage across the Atlantic 

 in Brande's Journal, vol. xiv. 



•j- The height of this mountain appears to be a matter of great uncertainty. 

 It would be interesting to know whether the isothermal lines would strike 

 it at the same height that they strike Chimborazo. Circumstances might be 

 mentioned which would have a tendency to depress, and others having a 

 tendency to elevate them. In calculating its height from the condition of 

 the mercury in the barometer on its summit, it is probable that the co- 

 efficient employed in Europe would be found inapplicable. 



% See Silliman's Journal, vol. ix. p. 4. 



§ See Von Buch on the Climate of the Canary Islands, in Edinburgh^ Phi- 

 losophical Journal for July 1826. 



