DYNAMICS OF CYCLONES AND ANTICYCLONES. 149 



and 1 cm. deep. This heated platform is placed on a table, and the boiler for 

 producing steam to heat it at some distance away, so as to prevent the heat of the 

 boiler producing disturbing currents near the experimental area. 



When we watch this hot area we will see the steam rising from its wet surface, 

 and according to the draughts in the room, or absence of them, it will either rise up 

 in an irregular manner, or be drifted across the hot area to the one side, but showing 

 in neither case any tendency to form cyclonic movements. Suppose the air in the 

 room to be perfectly still and the hot air ascending vertically. If we now blow gently 

 over one side of the hot area, a cyclone at once forms which collects the hot air into 

 its core and carries it to a considerable height. This cyclone soon dies away if we 

 cease to supply the tangential energy. 



Let us now see what the effect is if there is a draught across the hot area, and 

 we alter the conditions by placing a screen across the current so as to shield one 

 side of the hot area whilst it is allowed to blow over the other. A cyclone will now 

 at once be formed at a short distance from the edge of the screen — a sort of eddy, in 

 fact. It does not, however, remain in its place, but at once begins to travel across the 

 hot area, in the same direction as the cross air current, rotating in the direction 

 given by this tangential current, and rising to a considerable height above the 

 platform. 



It is not necessary that the screen protecting the one side of the platform 

 should have a vertical edge, though such an edge is best suited for making eddies. 

 The edge of the screen may be shaped like a magnified comb, 

 as shown in fig. 7. When so shaped the screen exerts very 

 little effect on the current at the tips of the teeth, but more 

 near their base, thus allowing a stronger current to blow over 

 one side of the area than over the other. If this screen be put 

 up across the current a cyclone will at once be seen forming 

 near the points, and so soon as formed it begins to travel away 

 across the hot area in the direction of the strongest tangential 



air current. Before this cyclone has gone far another forms near the screen, and it 

 in turn passes, following the course of the first, and so on, cyclone following cyclone 

 so long as there is a cross current and a heated area. 



It may be as well to point out here that though there may be a slight tendency 

 for eddies to form at the edge of the screen, in the above experiment, yet this 

 tendency is very slight and has but little to do with the formation of the cyclones. 

 If we repeat the experiment, without heating the experimental area, and using fumes 

 to show the movements of the air, it will be found that only very slight eddying 

 effects can be detected. Any that are formed with a screen having a vertical edge are 

 small, irregular and broken, and frequently cannot be traced at all, and when the comb- 

 like obstruction is used no eddying effect is produced unless there is an ascending 

 column of air. The question of the necessity of any eddying effect of the screen in 



