PHYSICS: L. PAGE 
47 
IS A MOVING STAR RETARDED BY THE REACTION OF ITS 
OWN RADIATION? 
By Leigh Page 
Sloane Physical Laboratory, Yale University 
Communicated by H. A. Bumstead, January 21, 1918 
A question of some interest to the astronomer is whether or not a body in 
motion, such as a star, is retarded by the reaction of its own radiation. For, 
on the electromagnetic theory of radiation as developed by Maxwell and his 
followers, a beam of radiant energy is supposed to have a quasi-momentum, 
such that if a body emits energy in a single direction it will lose momentum 
and in consequence suffer a reaction tending to push it in the opposite direc- 
tion. Now if a star is at rest, and in thermal equilibrium, it follows from sym- 
metry that it will radiate equally in all directions, and there will be no result- 
ant impulse. If, however, the star is in motion, classical electrodynamics 
leads to a greater emission in the forward direction than in the backward, 
and consequently it would appear at first sight as though there should be a 
retardation which would ultimately bring the star to rest. The problem 
has been treated in some detail by Professor Sir Joseph Larmor in the Pro- 
ceedings of the Fifth International Congress of Mathematicians, 1 held at Cam- 
bridge in 1912, and he finds the resistance to motion due to the radiation to be 
F = -vR/c 2 
where v is the velocity of the star, R the rate of emission of energy, and c the 
velocity of light. 
Apart from its intrinsic interest, Larmor's result is of importance in that it 
would constitute, if correct, a contradiction between classical electrody- 
namics and the Principle of Relativity (reference here is to the relativity of 
constant velocity systems, not to the broader conception of general relativity 
recently developed by Einstein). It is known, however, that the connection 
between classical electrodynamics and the Principle of Relativity is very 
close. Lorentz obtained the relativity transformations in his effort to give 
the electrodynamic equations for a moving system the same form as for a 
fixed system, even before Einstein advanced the relativity idea, and the 
author has shown that the electrodynamic equations can be derived in their 
entirety and exactly from the kinematical transformations of relativity and 
the assumption that each and every element of charge is a center of uniformly 
diverging tubes of strain. 2 
Now to calculate rigorously from the electrodynamic equations the reaction 
on so complicated a mass as a star would be hopelessly involved. Fortu- 
nately, however, the problem can be simplified to the extent of dealing with 
