Royal Society, 67 



and hence to find the potential at any point due to a given 

 distribution. For this purpose we may make an assumption, 

 viz. that a distribution of uniform density over a sphere must, 

 if that sphere be small enough and K be continuous, have 

 uniform potential over the sphere. If this be so, the above 

 process determines all the equipotential surfaces due to such a 

 distribution. 



But any electrified particle may be regarded as a distribu- 

 tion uniform in density over an infinitely small sphere. Hence 

 the equipotential surfaces due to any electrified particle can 

 be determined by the above process. Hence, jpurther, the 

 potential of any electrified s vstem can be determined, being 

 the sum of the potentials of its several parts. The inference 

 that we have drawn above respecting action at a distance does 

 not depend on this assumption respecting the potential of a 

 particle, and remains valid whatever be the true solution of 

 the problem of finding the attraction of a particle. 



IX, Proceedings of Learned Societies, 



ROYAL SOCIETY. 



[Continued from vol. iii. p. 538.] 



Feb. 22, 1877.— Dr. J. Dalton Hooker, C.B., President, in the 



Chair. 



THE following paper was read :— 

 " On Crookes's Porce." By G. Johnstone Stoney, M.A., P.B.S., 

 and Bichard J. Moss, B.C.S. 



In two papers by one o£ the authors of the present communi- 

 cation, which appeared in the Philosophical Magazine for March 

 and April 1876, it has been shown that the motion of the blackened 

 disks o£ a Crookes's radiometer can be explained by the known 

 dynamical properties of the trace of gas which is present, and the 

 term " Crookes's force " is proposed to designate the reaction 

 which comes into play between the blackened disks and the walls 

 of the exhausted chamber when a difference of temperature exists 

 between them. Shortly after the first of these papers appeared 

 we commenced an experimental investigation of the subject with 

 the view of learning, if possible, the laws to which the force 

 conforms. The investigation is still in progress ; and being 

 exceedingly tedious, it will require a great expenditure of time 

 before it is completed ; we propose, however, in this preliminary 

 paper to describe the apparatus and methods of observation em- 

 ployed, and to give some of the results already obtained. 



If the pressure which is exerted on the blackened pith surfaces 

 reacts on the sides of the glass envelope, it follows that a trans- 

 parent disk deUcately suspended close to a stationary disk of 



F 2 



