244 CARNEGIE INSTITUTION OF WASHINGTON. 



In the first trials difficulty was experienced from the light reflected diffusely 

 from the battery of small concave mirrors, but after these had been dis- 

 mounted, their surfaces retouched and covered with a "sputter" coat of plati- 

 num, much of the trouble disappeared. For future work, however, the 10 

 mirrors will be replaced by 2, the number of reflecting edges being thus 

 reduced in proportion. The last test of this arrangement was made on August 

 25, when the return image, with the mirror rotating, was found to be so bright 

 that it could be observed readily when the intensity was diminished tenfold. 



The present system of an air-blast acting upon paddle-wheels has not 

 proved sufficiently powerful with the air compressor now in use to give a 

 speed of 500 revolutions a second, and some modification will be introduced 

 before the work is continued next summer. The use of an electric motor, or 

 possibly some form of air turbine, should readily solve this difficulty. Experi- 

 ments will also be made with a view to obtaining a more steady arc lamp as 

 a source of light. 



With the definite knowledge that the intensity of the light returning from 

 the distant station will be amply sufficient for purposes of measurement, it 

 seems altogether probable that the method described will yield excellent 

 results. Michelson has, however, devised a means by which the original 

 Foucault arrangement can still be employed. A large number of small 

 plane mirrors are adjusted on the circumference of a circle of which the axis 

 of the rotating mirror is the center. The light from the image of the source 

 falls on the lower half of a face of the rotating mirror and is reflected to one 

 of the plane mirrors. Thence it is returned to the upper half of the rotating 

 mirror, from which it is reflected to the concave, and so to the distant station. 

 A succession of nearly parallel beams is formed as each of the small plane 

 mirrors comes into play. When the light from the distant station retraces 

 its path an image is formed which coincides with the source when the mirror 

 is either at rest or rotating so rapidly that an adjacent face replaces the original 

 one. As in the other form of apparatus, the return beam is observed with the 

 aid of a plane-parallel plate of glass. 



EFFECT OF THE EARTH'S ROTATION ON THE VELOCITY OF LIGHT. 



A brief description was given in last year's report of the experiment devised 

 by Professor Michelson for distinguishing between results which would follow, 

 on the one hand, from the generalized theory of relativity or the hypothesis 

 of a stationary ether, and on the other, from that of an ether either com- 

 pletely or partially dragged along by the earth in its rotation. The com- 

 puted value of the displacement of the interference fringes as required by the 

 theory of relativity should be about 0.15 fringe for a light circuit 1 mile in 

 length. From the results obtained during the summer of 1921, it was hoped 

 that the fringes would be sufficiently distinct to admit of accurate measure- 

 ment. 



Observations were made during the past summer over two circuits, one 

 approximately rectangular and 6,500 feet in length, the other triangular and 

 about 5,200 feet long. In both cases the fringes were found to be too unsteady 

 for measurement, although the observations were made in the early evening, 

 at the most favorable time for good conditions of seeing. Occasional improve- 

 ment was seen, but this was insufficient for the purpose in view. 



