200 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1921. 



This energy may still be perceptible to our eyes or apparatus when 

 reaching us from the stars after a journey which has consumed many 

 thousands of years. 



We know, too, that this energy, while it is on its way, travels in a 

 manner strikingly similar to the propagation of waves, so much so 

 that Ave feel justified in describing light as consisting of waves of 

 definite lengths and properties. 



Now how does this energy travel through apparently empty space 

 with these singular wave properties? The natural answer, almost 

 the intuitive answer, is to say that it travels through a medium, and 

 so we invent the " ether," simply as the medium which carries the 

 light. 



But if there is such a medium in space, and light travels through 

 it in every direction at the same speed, it would seem as if here, at 

 last, in this undisturbed ether, we had our frame of reference which 

 we could use as our basis for the measurement of all other motions. 



DETECTION AND MEASUREMENT OF MOTION BY LIGHT SIGNALS THROUGH 



THE ETHER. 



If this be true, we can detect whether this world of ours is moving 

 through the ether or not by sending light signals through equal dis- 

 tances in different directions and seeing whether they come back to us 

 at the same interval of time. 



To see how the thing works, let us suppose first that we have an 

 observer at rest with respect to the ether and surrounded by a circle 

 of mirrors set in various directions from him but all at a distance of 

 186,000 miles. 



If he then produces a flash of light at his own position this light 

 will travel out and in one second will reach all the mirrors simul- 

 taneously, will be reflected at each and at the end of another second 

 will come back to him simultaneously from all the mirrors. (If this 

 hypothetical apparatus appears to you inconveniently large, you can 

 just as well imagine one a million times smaller, which would make 

 the radius of the circle about a thousand feet, and count your time in 

 millionths of a second instead of whole seconds.) 



So far so good. But now suppose that the observer and his whole 

 circle of mirrors, big or small, are not at rest but are all moving 

 together uniformly at a speed of half the velocity of light. 



Now let the observer send out a light signal and wait for its reflec- 

 tion from that mirror which is directly on the line of his track and 

 in the direction toward which he is moving. 



The light traveling out toward this mirror would itself move 

 186,000 miles a second but would have a "stern chase," since the 



