522 



Mr. H. S. Allen on the Motion of 



revolving dish H, driven by the Froment's engine F, and 

 then in succession through the cardboard shade G, the 

 Thornton-Pickard shutter B, the glass vessel 0, and a rect- 

 angular aperture in the screen B on to the plate at A. 



Fig. 1. 



Method of Release. 



In order to obtain undisturbed motion through the liquid, 

 it is clear that the sphere must be released beneath the sur- 

 face. Worthington and Cole* have shown that when a 

 sphere is allowed to fall from air into water, a disturbance of 

 a large body of water takes place and the sphere carries 

 down with it a bubble of air. It therefore became necessary 

 to devise some method of instantaneously releasing the ball 

 under water without giving rise to any appreciable disturbance 

 of the liquid, and without communicating any velocity of 

 translation or rotation to the ball. After trying some 

 mechanical devices, it appeared that by far the simplest 

 method was to suspend the spheres by a small straight 

 electromagnet. As this involved the use of magnetic material 

 for the spheres, it was decided to experiment with bicycle 

 bearing-balls. These could be obtained in six different sizes, 

 with diameters ranging from about # 3 centim. to 0*8 centim., 

 and careful measurements showed extremely small divergence 

 from true sphericity. 



The following sketch' shows diagrammatically the method 

 employed for releasing the shutter and the falling weight 

 which gives the plate its horizontal motion, at appropriate 

 instants of time after the ball to be photographed has com- 

 menced falling. The release is effected by a second ball 

 dropping in another vessel on to a light platform of aluminium 



* Phil. Trans, clxxxix. (A), pp. 137-148 (1897). 



