It is evident that the velocity of the wind in a revolving storm must be the greatest and the least respectively on opposite sides of the axis of rotation, in a diameter which is perpendicular to the path of that axis; for on one side the direction of the revolving current conspires with that of the progressive motion of the storm, and on the other it is con trary to it. In other parts within the limits of the storm the direction and velocity of the wind must be compounded of the rotative and progressive motions; and it will happen frequently that a temporary calm is experienced at each point on the earth's surface at which the axis of the storm successively arrives.
The phenomena of tropical storms are not precisely such as they would be if the air had a simple movement of rotation ; the particles of air, while revolving, are probably subject to undulatory motions in spiral curves, and the wind appears sometimes to shift to different points all round the compass. 31r. Redfield states that, in small whirl winds, the axis of rotation appears at times to describe gyrations In looped curves about Its mean place in the line of progressive motion ; and the like gyrations probably take place in those of an extensite kind ; indeed in some voyages under the Influence of revolving storms they have been actually observed : but in order to simplify the expla nation of the phenomena of whirlwinds, it is usual to assume that the particles of air revolve In the circumferences of circles whose centres are in the axis; the latter having at the name time a movement of progression in a rectilinear or curvilinear direction. Now, if the plane of the paper represent the surface of the sea, and a line through A„ perpendicular to it, represent the axis of a whirlwind whose north and south diameter is N., S., and in which the particles of air are sup posed to revolve (for example) in the direction indicated by the order of the letters N., W., S., E. ; the progressive movement of the axis being also supposed to be from A, through N. or from south to north: thou, since at N. a tangent to the circle lies N., east and west, it is evident that a ship at that point would experience a wind blowing from the met when the centre of the storm is at ; and if the ship remain stationary, the wind will continue to blow from the same quarter till arrives 4t N., the tangents to the concentric) circles supposed to be described by the partieles being due east and west at the northern points of the circumferences as they eueessively arrive at N , and the wind in all the northern half of the Storm revolving in the direction E., N., W.: but after this time, the wind blowing in the direction W., S., E., must be felt at N. as dwest wind till tharemaining
half of the storm has passed over that point. In like manner, if the axis of the storm were to move from A towards W'., a ship supposed to be stationary at the latter point would feel the gale from the north till A, arrives at W. ; after which, as the eastern semicircle passes over that point, the ship would experience a wind from the south.
Again, if the axis were to move from A, towards that is, from south-west to north-east, for example, the direction of the whirlwind being, as before, according to the order of the lettere N., W., S., E., and the ship being supposed to remain stationary at some point, as al, till the storm has passed over it, then the lino of direction iu which the points of the whirlwind successively overtake the ship being 11, al, parallel to A, the arcs sat„ m„ &c., will represent the several directions in which the wind will successively be felt at the ship during the continuance of the storm. Thus, the axis of the whirlwind being at A„ the convex surface of the storm has just reached the ship, and the wind blows in the direction a m„ or iu the circumference of the circle whose centre is A, that is, nearly from the eastsouth-east ; next, the axis being at the point at, in the circumference of the circle whose radius is A, is at It, ; and then, at the ship the wind is felt in the direction Ent„ or in the circumference whose radius is A, „ or its equal A that is, nearly from the east-by-south. Again, the axis being at the point in the circumference whose radius is A, 11,, ill at at, ; and then at the ship the wind is felt in the direction em,, m in the circumference whose radius is A, m„ or its equal that is, from the north-east. When the axis is at A, and A, the points st, and sr, arrive at at,, and the wind there is felt successively in the directions dm, and em„ that is, nearly from the north-by-west, and from the north-north-west. When the axis has advanood beyond it is evident that the whirlwind ceases to have any effect on a ship at It,. If tan gents were drawn at m, to the arcs bas„ Itc., they would evidently be parallel to tangents at the corresponding points m.„ Ite.; there fore the directions in which the circumference of the concentric circles meet the line of direction al, al, will be those in which the wind is felt at the ship during the storm. In like mauner, the successive direc tions in which the wind blows in a revolving storm may be exhibited, whatever be the situation of the ship and the movement of the axis of rotation.