344 Dr. W. F. G. Swann on th 



It' we set a piece o£ matter in motion in a straight line 

 with constant velocity, the electromagnetic theory leads to 

 the conclusion that whatever may be the nature of the 

 internal electronic, atomic, and molecular motions, the 



system as a whole will contract in the ratio ( 1 — ,-^ I 



in the direction of motion in such a way to insure that at 

 corresponding times the x component of any electron 



1 — p2 ) times the x component of the electron 



in the fixed system *. Of course, so far as we are concerned 

 in our notions of the constitution of matter as founded on the 

 electromagnetic theory, all differences of elnstic properties, 

 &c. are to be attributed to differences of the strengths, 

 distribution, and motions of the electrons. The curious 

 thing is that the above contraction is absolutely independent 

 of the nature of the elastic properties of the matter con- 

 sidered, i. e. t it is independent of the nature of the electrons 

 and of their motions in the system f . Now when we come 

 to consider problems of motion other than motion in a 

 straight line with uniform velocity, it may be, and in fact 

 certainly will be, that the final condition of affairs will 

 depend not only on the motion imparted to the system but 

 also on the motions which the electrons had in the system 

 when at rest. 



It is interesting to attempt to treat the problem of uniform 

 rotation on the principle adopted by Sir Joseph Larmor for 

 uniform translatory motion; for in spite of the fact 

 that the principle of relativity is a principle which is 

 postulated independently of its complete verification from 

 the electromagnetic theory, it must nevertheless be looked 

 upon as being suggested by direct argument, and it is of 

 interest to trace the course of an analogous line of argument 



CD O 



for the ca?e of uniform rotation. 



Let us first briefly review the problem of uniform 

 translatory motion in the manner given by Larmor (' iEther 

 and Matter/ pages 167-177). 



* Strictly speaking this is not true : tlie electromagnetic theory 

 really only leads to this conclusion when the system discussed is one in 

 which all the electrons in the fixed system are absolutely devoid of 

 motion (see final paragraph of this paper). 



t Really this fact is not as carious as it seems when we realize that 

 the conclusion as derived directly from the electromagnetic theory only 

 strictly holds for one particular kind of system, viz., one in which the 

 electrons have no orbital motions. It is the postulation of the principle 

 as holding for all systems which makes the result seem so startling. 



