2 THE ROYAL SOCIETY OF CANADA 



the majestical roof fretted with golden fire," so that the "floor of 

 heaven was thick in-laid with pattens of fine gold." Beneath the 

 earth, the abode of demons ! 



This idea was followed by a round but stable earth around which 

 the sun, moon, planets and stars rotated. A relative velocity of 

 rotation with a vengeance! 



The heliocentric theory perhaps assumed a stable sun, almost 

 certainly fixed stars. To-day every particle of matter moves. There 

 is neither absolute place, nor time, nor velocity, and we are led to 

 expect that the quest for them is as futile as perpetual motion. 



One hope further did until recently remain. If aether was a 

 compacted plenum filling all space, then whether material or electro- 

 magnetic it might afford a standard with reference to which absolute 

 velocities could be obtained. 



This idea was dominant in many able minds. It was destroyed 

 by experiments, such as that of Michelson and Morley. 



The explanation of Fitzgerald, worked out by Larmor and 

 Lorentz, that matter no less than electrons was flattened by velocity 

 through space, suffices to account for the null experiments. Physicists 

 were content with so moderate a demand, involving with the earth's 

 speed, but a few inches in its diameter of 8,000 miles. The aether 

 survived, and it might yet be stagnant! 



In 1905 Einstein accepted the transformation equations of 

 Larmor and Lorentz, but enunciated a great principle. 



All linear motions are independent of any frame of reference. 

 Light travels at the same speed to all observers, and is independent of 

 the velocity of the source. This broad generalization was incapable 

 of verification, but the wideness of its scope brought the principle into 

 wide-spread favour. Incidentally some curious results followed from 

 the general position now achieved. Two bodies or waves approaching 

 each with velocity of half that of light would come together with a 

 velocity apparent to an observer with each, not as the velocity of 

 light but as 4/5 of the velocity of light. If each moved towards the 

 other with velocity c, relative to some object, then their relative 

 velocity to a massless traveller with either would still be c and not 

 2c as ordinary kinematics might suggest. The Fitzgerald shorten- 

 ing still held good, and the transformation equations of Larmor and 

 Lorentz were consistent in their application to Maxwell's Electro- 

 magnetic equations and to leading phenomena, such as Doppler's 

 effect, pressure of light, etc., resulting from the equations. 



But if Einstein's principle of relativity was accepted for linear 

 velocities, it was impossible for linear acceleration, which is merely 



