218 Messrs. Searle, Aldis, and Dobson on Revolving Table 



infinite perfect fluid will, under similar conditions. To 

 produce any variation in the speed or direction of motion o£ 

 an electron, other electrified particles must be present to 

 exert an external force upon it. Thus Kaufmann determined 

 m" by measuring the deviation in path of an electrified 

 particle which moved parallel to an electrified plane surface. 

 But if we now call m the true mass and m" the apparent, or 

 hydrodynamic, mass of a ball moving in a perfect fluid and 

 parallel to a fixed plane surface, can we not also separate 

 these two masses ? The formula for the kinetic energy of a 

 sphere, of density p and radius a, moving in a perfect fluid 

 parallel to, and at a distance h from, a fixed plane surface, is * 



We also have 



Since the hydrodynamic mass, m", is a function of the 

 variable h and the true mass, m , is a constant, we may 

 measure the kinetic energy of similar spheres which move 

 parallel to a plane boundary but at different distances from 

 it and so separate m Q and m". Practically, we should ex- 

 perience great difficulties ; the effect would be very small ; 

 we have no perfect fluids; and we have not yet deduced the 

 equations for spheres moving in a viscous fluid. But these 

 are purely experimental difficulties and show no essential 

 differences between the conceptions we should make for 

 hydrodynamic and electromagnetic mass. Both may be 

 considered as variable quantities of the same character added 

 to the true and constant inertia of matter. 



University of Cincinnati, 

 August, 1910. 



XXV. On a Revolving 7 able Method of determining the 

 Curvature of Spherical Surfaces. By G. F. 0. Searle, 

 M.A., F.R.S., A. C. W. Aldis, M.A., and G. M. B. 

 Dobson, B.A.f 



§ 1. lyiOR many optical purposes it is necessary to 



J? determine the radius of curvature of a convex 



or a concave surface with considerable accuracy. The 



revolving table method, described below, has the advantage 



that the radius is found directly from two readings on a 



* Lamb, ' Hydrodynamics,' p. 143. 

 f Communicated by the Authors. 



