W.

b. If on the other hand the latitude, and consequently the velocity of rotation of the surface with which the air is in contact, be not constant, the vane can preserve its direction unchanged, solely on the condition that the interval between the extreme points of its course remain unchanged, and at the same time its velocity do not vary. The cases in which the vane changes its direction are also three in number:

WNW.

WSW.

1. When the point towards which the wind is moving,

or from whence it has

started, changes its position. If (fig. 1) we suppose this point to move from SW. to

NE., we find that at the centre of the circle

the wind chops about, after a calm, from NE. to SW.; on the right hand side of the path of the centre it veers from ENE. through E.

to SW., and on the left hand side of the path, from NNE. through W. to WSW. In each of these cases the amount of change diminishes as the distance from the centre increases. The same reasoning holds for winds which diverge from one central point, if we bear in mind that the directions of the wind in this case are diametrically opposite to those in the case last considered.

2. When the point around which the air rotates changes its position. If the mass of air, rotating from right to left, in a direction opposite to that of the hands of a watch, move from SW. to NE., the place of observation will pass successively through the points of intersec

tion of a right line, with the concentrical circles of the cyclone passing over it. In this case the directions of the vane represent the

tangents to these circles drawn at the points of intersection. At the centre the wind chops about from SE. to NW.; on the right hand side it veers from SSE. through S. towards WNW., and on the left hand side from ESE. through E. towards NNW. The

direction in which the change of direction takes place, on either side of the path, is the same as in the former case; but the directions of the wind differ by an angle of 90° in the two cases. If the cyclone rotate from left to right, the directions of the wind will be exactly reversed.

3. When, in the case of a constant wind, the distance of the initial point or the velocity of the storm is altered.

The effect produced by a change of position of the initial point of the current may be perceived from the following consideration :

The velocity of rotation of the individual points on the earth's surface varies as the diameter of their latitude, and therefore increases from zero at the poles to a maximum at the equator. If the air be at rest, it partakes of the velocity of rotation of that portion of the globe over which it rests. Hence, if a difference of temperature, or any other cause, impart to it a tendency to

move in a direction parallel to the equator, its motion cannot be affected by the rotation of the earth, inasmuch as the points on the earth's surface towards which it is moving have exactly the same velocity of rotation as those which it has left. If, on the other hand, the air be set in motion by any cause from the poles to the equator, it passes from points which possess a lesser to those which possess a greater velocity of rotation. This

air, therefore, moves at a slower rate from W. to E. than the points with which it comes in contact; and hence it appears to have a motion in the opposite direction, i.e. from E. to W. The deflection of the wind from its original direction, its rate of progression being supposed unaltered, will be greater, the greater the difference of the velocities of rotation of the point from whence it has started, and of that where the observation is made; or, in other words, the greater the difference of latitude of these two places. Hence we see

1. In the northern hemisphere winds, originally north winds, change gradually through NE. towards E. If we take points

so situated that A B C D are all under the same meridian, A being the most northerly, D the most southerly station; and that A A,, &c., lie on the same parallel of latitude, A being the most westerly, and A,,, the most easterly. If we conceive the whole body of air lying between A A,,, and D D,,, set in motion from north to south by any cause, we shall find that the air from C C reaches the latitude DD,,, more or less as a north wind;

that from B B as a north-east wind; while that from A A, will reach it still more as an east wind. In this case an observer in the latitude DD, will see the vane gradually move from N. through NE. towards E.

2. In the southern hemisphere, winds, originally south winds, change gradually through SE. towards E. If we take, as before, points

d d, d,, d

с с C11 C bb, b, b

a a, a, a

of which those lying in the latitude a a,,, are the most southerly, and those in the latitude d d,,, the most northerly, an observer in the latitude dd,,, will see the vane gradually turn from S. through SE. towards E.

If in either hemisphere an east wind has arisen in the manner above indicated, it will preserve its motion in the latitudes DD,,, or dd,,, without being in any way affected by the rotation of the earth.*

If the cause which produced a current towards the equator continue in operation, the east wind, which has thence arisen, will interfere with and retard the current. By means of this retardation, the air will soon attain the velocity of rotation of the places with which it is in

*

Ferrel, The Motions of Fluids and Solids relative to the Earth's Surface; comprising Applications to the Winds and the Currents of the Ocean. New York, 1860, p. 25. 'In whatever direction a body moves on the surface of the earth, there is a force arising from the earth's rotation which deflects it to the right in the northern hemisphere, but to the left in the southern.' This extension of the principle of Hadley's theory leads to the same conclusions as that theory in the case of all the phenomena which are here considered. If we take into consideration the components of the moving force, which result for motions in the direction of the parallels of latitude, the conclusion which has been drawn as to the change of direction of the wind holds good for all winds, including east and west.

contact, and will pass into a state of relative rest with regard to it. A constant tendency of the air to flow towards the equator will therefore reproduce precisely the phenomena which have been just considered.

If we now take the case of the appearance of equatorial currents, after polar currents have predominated for a time in the northern hemisphere the incipient equatorial current, as a south wind, will displace the polar current, which has become more or less easterly, in the direction E., SE., S.; in the southern hemisphere, as a north wind, it will displace the polar current, which has become easterly like the other, in the direction E., NE., N.

Hence, on the whole, in the latitude DD,,,, in the northern hemisphere, the observed change of the wind will be N., NE., E., SE., S.

In the latitude dd, in the southern hemisphere, it will be exactly the reverse, viz., S., SE., E., NE., N.

Air, which flows from the equator towards the poles, moves from points which possess a greater to those which possess a lesser velocity of rotation towards the east. Hence it follows that

3. In the northern hemisphere a southerly wind in its progress gradually veers through SW. towards W.

4. In the southern hemisphere a northerly wind in its progress gradually veers through NW. towards W. If we take points

in the northern hemisphere, such that those in the latitude G G are the most southerly, we shall see, as before,