202 Prof. J. D. Everett on the General Circulation 



fore having an excess of centrifugal force. This excess can be 

 resolved into two components, one of which is vertically upwards, 

 and produces an inappreciable diminution of barometric pressure 

 under it. The other component is directed along the meridian 

 towards the equator, and will produce a deflection towards the 

 equator unless the air which lies on that side has greater elastic 

 force than the air which lies on the polar side. This component, 

 like the vertical one, would be inappreciable if it only acted 

 through a belt five miles wide ; but when the air over a zone 

 many degrees wide has a mean motion from the west, the ac- 

 cumulated effect, in the shape of differential pressure at its two 

 margins, is very considerable. 



These remarks respecting the excess of centrifugal force pos- 

 sessed by west winds, apply equally to the defect of centrifugal 

 force in east winds, except that the pressure which they exert in 

 virtue of their relative velocity is directed/rom the equator instead 

 of towards it. 



Excess and defect of centrifugal force thus produce an increase 

 of barometric pressure from the polar to the tropical limit of the 

 region of west winds, and a similar increase from the equatorial 

 to the tropical limit of the region of east winds. That is to say, 

 the effect due to centrifugal force is everywhere of the same kind 

 as the observed difference*. 



As regards the causes of the east and west winds, the prime 

 mover is the increase of temperature from the poles to the 

 equator, which would of itself give a north wind over the whole 

 of the northern hemisphere, and a south wind over the whole of 

 the southern, at the earth's surface, with return-currents in the 

 opposite directions above; but in virtue of the earth's rotation, 

 which is carrying all points of its surface from west to east with 

 velocities proportional to their distances from the axis, a body 

 set in motion along a meridian towards the equator will fall 

 behind the meridian unless constantly subjected to differential 

 pressure on its two sides ; and a body moving from the equator 

 towards either pole tends, in like manner, to move in advance 

 of the meridian along which it is travelling. Most writers, in 



* It is necessary to remark, by way of caution, that the effect in ques- 

 tion depends upon the movement of the whole body of air over a region^ 

 and not merely of the lower portion whose motion constitutes our observed 

 winds. If some strata are moving to the east and others to the west, op- 

 posite signs must be given to eastward and westward velocities, and the 

 average of the whole (not a height-average, but a mass-average) must be 

 struck. We assume that this average velocity is westward in the trade- 

 wind regions and eastward in the temperate zones, thus agreeing with the 

 observed winds. Indeed the fact of a continual circulation of air between 

 the polar and equatorial regions, seems to involve, as a result, that the in- 

 tertropical parts of the earth's surface are revolving faster, and the other 

 partg pf the earth's surface slower than the atmosphere over them. 



