156 Dr. G. T. Walker on 



emitted in different directions from a point on the earth's, 

 surface. 



Suppose that an observer B at the time t = emits a flash 

 of light from S (fig. 1) to a point P on the earth at a 

 distance of I km. from S. The time T consumed by the 

 light in going from S to P and back to P is accord- 

 ing to (1), 



I=f( 



1 + 



B's watch can consequently be adjusted by means of 

 light-signals without contradiction. 

 Trondhjem, Norway, 

 June, 1918. 



XIII. On Fermat's Law. 



India Meteorological Department, 

 Simla, 16th April, 1918. 



To the Editors of the Philosophical Magazine. 

 Gentlemen, — 



IN a paper on Format's law by Professor D. N. Mallik, in 

 your issue of July 1913, he deduces (p. 152) that 

 " optical energy is entirely kinetic," and hence that the 

 phenomena of elasticity and electrostatics are also kinetic 

 These results, if valid, are of very great generality and 

 importance, and they have been restated on pp. 12 and 13 of 

 Professor Mallik's recent volume on 'Optical Theories'*.. 

 The idea that all potential energy is capable of interpreta- 

 tion as kinetic was worked out by Helmholtz, and J. J. 

 Thomson has examined many of its consequences ; but the 

 conclusion that optical energy must be kinetic is, I believe, 

 entirely new, and as I do not follow the argument employed, 

 I desire to append the following criticism in the hope that 

 a decision may be reached in a matter so far-reaching in its 

 consequences. 



2. Professor Mallik says on page 149 : — 



" the configuration of equilibrium and motion of a dy- 

 namical system is defined by 8\(T — Y)dt = 0, where 



T= kinetic energy ? 

 Y = potential energy. 



* Cambridge University Press, 1917. 



