RELATIVITY — RUSSELL. 201 



mirror is receding half as fast as it is traveling, and it is easy to see 

 that it would take two whole seconds to reach the mirror. 



On the return journey the observer will be advancing to meet it 



with half the speed of light, and this part of the process will take 



only two-thirds of a second. The elapsed time for the round trip of 



the light will be two and two-thirds seconds, considerably longer than 



if the observer was at rest. 



Consider next a ray of light which gets reflected in the mirror 

 whose direction from the observer is at right angles to the first. 



It will not have the long stern chase which the first ray has, but 

 nevertheless it will lose something, because in order to reach the 

 moving mirror it will have to travel " on the bias," so to speak, 

 through space, so that it will reach not the point where the mirror 

 was when the light started, but the point where it will be when it 

 gets there, and something quite similar will happen on the return 

 journey 



When this is calculated it is found that the round trip will in this 

 case take about two and one-third seconds. (The exact amount in- 

 volves calculating a square root that we need not bother with here.) 



The important point is that in this case, where the observer and 

 mirrors are moving through the ether, the ray of light which has 

 traveled up and down the direction of motion will take a longer time 

 for the round trip than the ray which has traveled crosswise to the 

 motion over a path of exactly the same length. 



We should, therefore, in this way be able to detect motion of our 

 own system through the ether, and if our measurements were suffi- 

 ciently accurate, determine its direction and rate. 



FAILURE OF EARLY EXPERIMENTS. 



This was attempted in the famous Michelson-Morley experiment. 

 The distance of the round trip was in this case only a few feet, and 

 the difference in time over the two paths only something like a 

 millionth part of one billionth of a second. 



But this minute interval could be measured by splitting a ray of 

 light into two parts by letting part of it be reflected sidewise from a 

 transparent mirror and the rest go through, and reuniting the parts 

 after their trip. 



If one had gained on the other by even a fraction of the time of 

 vibration of a single light wave the fact could be detected, and the 

 waves which we ordinarily call light vibrate at the rate of about 

 six hundred thousand billion per second. 



Michelson and Morley tried their experiment, and in place of the 

 easily measurable result which they anticipated, they got nothing. 



