182 
Proceedings of the Royal Society of Edinburgh. [Sess. 
collected in a small cyclone of about 2 or 3 cm. diameter. It will also be 
seen that in this case, as in the water cyclones, the diameter is very small, 
and that the low-pressure area into which it is drawn is large — very many 
times larger than the cyclone. The high angular velocity of the air sur- 
rounding the centre makes it act like a solid tube, through which the 
steam is drawn. 
The general appearance of these suctionally-driven cyclones in water 
and air, as shown in figs. 1 and 2, presents a certain likeness to water- 
spouts, and possibly to tornadoes ; and if we could only discover the 
existence of sufficiently low-pressure areas over them, we might then 
be a little nearer to understanding their formation. 
It is well known that rapid motion confers on chains and bands certain 
properties which we associate with solid bodies. If a long loop of chain 
be hung over a pulley and driven at a great velocity, the chain behaves 
very much as if it were a solid at rest. If we strike it a blow anywhere 
below the pulley, we simply make a dent at the place. The chain acts 
very much as if it were a bar of lead ; only, the dent made by the blow 
will slowly straighten out owing to the friction of the links, and other 
causes, the chain behaving very much as if it had viscid properties. If 
the loop of chain be put on a pulley large enough to give it nearly a 
circular form, and set in rapid motion, and while still moving dropped off 
the pulley, it will run along like a hoop and go many yards before its 
motion is spent and it collapses on the floor. Cyclonic motion seems to 
confer on water and gases similar properties, giving them a certain degree 
of rigidity which they otherwise do not possess. It is well known that 
vortex rings vibrate after encountering an obstacle which tends to deform 
them. Or if we make the rings oval, they are seen to vibrate ; their long 
diameter, if originally horizontal, changes to a vertical position, and then 
back to the horizontal, and continues to oscillate between the two positions 
till they end their existence. They act very much like an elastic solid. 
The cyclonic movements in water and air (figs. 1 and 2) have also acquired 
properties associated with solids; they tend to keep straight and develop 
in a direction at right angles to the plane of the rotational component of 
the spiral movement in them. This is well seen in the case illustrated in 
fig. 2. Here the centre of the cyclone has developed in a direction at 
right angles to the wet surface which determined the plane of its rota- 
tional movement. It will be noticed that the cyclone is uninfluenced as 
regards straightness and direction by the movements of the surrounding 
air, till it meets with the strong cross current going up the chimney, where 
it gets broken up, as cyclones seem to have a distinct objection to bending. 
