374 A CENTURY OF SCIENCE 



tute at Berlin. So sensitive was the instrument to outside 

 vibrations that even after midnight it was found impos- 

 sible to get consistent readings. Finally a satisfactory 

 foundation was constructed in the cellar of the Astro- 

 physical observatory at Potsdam. But what was the 

 astonishment of the experimenters to find that the 

 expected shift of the interference fringes did not exist! 



The extreme delicacy of the experiment made it desir- 

 able to confirm the result by repeating it. This was 

 done by Michelson and Morley (34, 333, 1887) in 1887. 

 In place of a revolving table a massive slab of stone 

 floating on mercury was used to carry the apparatus. 

 This slab was kept in constant rotation, the observer 

 following it around. Moreover, the precision of the 

 experiment was greatly increased by reflecting each ray 

 back and forth across the slab a number of times between 

 leaving and returning to the mirror m. The accuracy 

 attained was such as to justify Michelson in declaring 

 that if the effect sought actually existed it could not be 

 so great as one-twentieth of its calculated value. In 

 1905 Morley and Miller 13 repeated the experiment for the 

 second time and succeeded in increasing the sensitiveness 

 of the apparatus to a point such that a motion through 

 the ether of one-tenth of the earth's orbital velocity 

 could have been detected. 



The displacement looked for in the Michelson-Morley 

 experiment is known as a second-order effect in that it 

 depends upon the square of the ratio of the velocity of the 

 earth to that of light. Michelson at first considered that 

 the negative result obtained confirmed a theory proposed 

 by Stokes in which it was assumed that the ether inside 

 and near its surface partakes of the motion of the earth, 

 while that at a distance is practically quiescent. But 

 there are many objections to Stokes' theory, one of which 

 was brought out by an experiment of Michelson 's (3, 475, 

 1897) in which he attempted by an interference method 

 to detect a difference in the velocity of light at different 

 levels above the earth's surface. The negative result 

 obtained led him to conclude that if Stokes' theory were 

 true the earth's influence on the ether would have to 

 extend to a distance above its surface comparable with 

 its diameter. Meanwhile a more satisfactory explana- 



