Such effects seem to show a rising current of air.
I will endeavour to explain my conceptions of this matter by figures, representing a plan and an elevation of a spout or whirlwind.
I would only first beg to be allowed two or three positions mentioned in my former paper.
1. That the lower region of air is often more heated, and so more rarefied, than the upper; consequently, specifically lighter. The coldness of the upper region is manifested by the hail which sometimes falls from it in a hot day.
2. That heated air may be very moist, and yet the moisture so equally diffused and rarefied as not to be visible till colder air mixes with it, when it condenses and becomes visible. Thus our breath, invisible in summer, becomes visible in winter.
Now let us suppose a tract of land or sea, of perhaps sixty miles square, unscreened by clouds and unfanned by winds during great part of a summer's day, or, it may be, for several days successively, till it is violently heated, together with the lower region of air in contact with it, so that the said lower air becomes specifically lighter than the superincumbent higher region of the atmosphere in which the clouds commonly float: let us suppose, also, that the air surrounding this tract has not been so much heated during those days, and, therefore, remains heavier. The consequence of this should be, as I conceive, that the heated lighter air, being pressed on all sides, must ascend, and the heavier descend; and as this rising cannot be in all parts, or the whole area of the tract at once, for that would leave too extensive a vacuum, the rising will begin precisely in that column that happens to be the lightest or most rarefied; and the warm air will flow horizontally from all points to this column, where the several currents meeting, and joining to rise, a whirl is naturally formed, in the same manner as a whirl is formed in the tub of water, by the descending fluid flowing from all sides of the tub to the hole in the centre.
And as the several currents arrive at this central rising column with a considerable degree of horizontal motion, they cannot suddenly change it to a vertical motion; therefore, as they gradually, in approaching the whirl, decline from right curved or circular lines, so, having joined the whirl, they ascend by a spiral motion, in the same manner as the water descends spirally through the hole in the tub before mentioned.
Lastly, as the lower air, and nearest the surface, is most rarefied by the heat of the sun, that air is most acted on by the pressure of the surrounding cold and heavy air, which is to take its place; consequently, its motion towards the whirl is swiftest, and so the force of the lower part of the whirl or trump strongest, and the centrifugal force of its particles greatest; and hence the vacuum round the axis of the whirl should be greatest near the earth or sea, and be gradually diminished as it approaches the region of the clouds, till it ends in a point, as at P, Fig. 2. in the plate, forming a long and sharp cone.
In figure 1, which is a plan or groundplat of a whirlwind, the circle V represents the central vacuum.
Between a a a a and b b b b I suppose a body of air, condensed strongly by the pressure of the currents moving towards it from all sides without, and by its centrifugal force from within, moving round with prodigious swiftness (having, as it were, the entire momenta of all the currents → → united in itself), and with a power equal to its swiftness and density.