798 Professor Sir J. J. Thomson [March 18, 



downwards as in Fig:. 4 ; this makes the ball fall more quickly than it 

 otherwise would, and thus tends to prevent it going out of the court. 



Before proceeding to the explanation of this effect of spin I will 

 show some experiments which illustrate the point we are considering. 

 As the forces acting on the ball depend on the relative motion of the 

 ball and the air, they will not be altered by superposing the same 

 velocity on the air and the ball : thus, suppose the ball is rushing 

 forward through tlie air with the velocity V, the forces will be the same 

 if we superpose on both air and ball a velocity equal and opposite to that 

 of the ball ; the effect of this is to reduce the centre of the ball to rest, 

 but to make the air rush past the ball as a wind moving with the 

 velocity V. Thus, the forces are the same when the ball is moving 

 and the air at rest, or when the ball is at rest and the air moving. 

 In lecture experiments it is not convenient to have the ball flying 

 about the room, it is much more convenient to keep the ball still and 

 make the air move. 



The first experiment I shall try is one made by Magnus in 1852 ; 

 its object is to show that a rotating body moving relatively to the air 



Fig 



is acted on by a force in the direction in which the nose of the body 

 is moving relatively to its centre : the direction of this force is thus 

 at right angles, both to the direction in which the centre of the body 

 is moving, and also to the axis about which the body is spinning. 

 For this purpose a cylinder A (Fig. 7) is mounted on bearings so that it 

 can be spun rapidly about a vertical axis ; the cylinder is attached to 

 one end of the beam B, which is weighted at the other end, so that 

 when the beam is suspended by a wire it takes up a horizontal posi- 

 tion. The beam yields readily to any horizontal force, so that if the 

 cylinder is acted on by such a force, this will be indicated by the 

 motion of the beam. In front of the cylinder there is a pipe D, 

 through which a rotating fan driven by an electric motor sends a blast 

 of air which can be directed against the cylinder. I adjust the beam 

 and the beam carrying the cylinder, so that the blast of air strikes the 

 cylinder symmetrically ; in this case, when the cy Under is not rotating 

 the impact against it of the stream of air does not give rise to any 

 motion of the beam. I now spin thecyhnder, and you see that when 

 the blast strikes against it the beam moves off sideways. It goes off 



