THE THEORY OF RELATIVITY 435 



In the first place then, how has it come about that our conceptions 

 of the fundamental notions of mechanics have been proved wanting? 

 This crime like many another may safely be laid at the door of the 

 physicists, those restless beings who, with their eternal experimenting, 

 are continually raising disturbing ghosts, and then frantically imploring 

 the aid of the mathematicians in order to exorcise them. Let us briefly 

 consider the experiment which led us into those difficulties from which 

 the principle of relativity alone apparently can extricate us. 



Consider a source of sound A at rest (Fig. 1), and surrounded by 

 air, in which sound is propagated, also at rest. Now, as every schoolboy 

 knows, the time taken for sound to go to B is the same as that taken to 

 go to C, if B and C are at the same distance from A. The same is true 



> 



~ B A B 



Fig. 1. Fig. 2. 



also if A, B and C are all moving with uniform velocity in any direction, 

 carrying the air with them. This may be realized by a closed railway 

 car or a boat. But if the points A, B, and C are moving with uniform 

 velocity, and the air is at rest relative to them, or what is the same 

 thing, if they are at rest and the air is moving past them with uniform 

 velocity, the state of affairs is very different. If the three points are 

 moving in the direction indicated by the arrow (Fig. 2), and if the air 

 is at rest, and if a sound wave is sent out from A, then the time required 

 for this sound wave to go from A to C is not the same as that required 

 to go from A to B. Now as sound is propagated in air, so is light in an 

 imaginary medium, the ether. Moreover, this ether is stationary, as 

 many experiments show, and the earth is moving through it, in its path 

 around the sun with a considerable velocity. Therefore we have exactly 

 the same case as before, and it should be very easy to show that the 

 velocity of light in a direction perpendicular to the earth's direction of 

 motion is different from that in a direction which coincides with it. 

 But a famous experiment of Michelson and Morley, carried out with the 

 utmost precision, showed not the slightest difference in these velocities. 

 So fundamental are these two simple experimental facts, that it will be 

 worth while to repeat them in slightly different form. If the three 



