of the vEther hi an Electromagnetic Field. 21 



The method adopted for the detection and measurement of 

 this motion was similar to that used by Professor Oliver 

 Lodge*. Three small plane mirrors are placed vertically at 

 three corners of a square marked out on a rigid table. These 

 mirrors were mounted in such a way that, by the pressure of 

 screws behind, a slight rotation round a vertical or round a 

 horizontal axis could be given to each. The mirrors were 

 placed with their planes perpendicular to the diagonals of the 

 square. At the fourth corner and with its plane in the dia- 

 gonal there was placed a fourth mirror, vertical as the others, 

 but " silvered " only to such an extent that a beam of light 

 incident at 45° is divided into two beams of approximately 

 equal intensity, one transmitted, the other reflected. 



Fiff. 1. 



A beam of parallel light from the collimator C (fig. 1) falls 

 on this mirror (a) at the required angle of 45°. The trans- 

 mitted beam passes along the side ah, while the reflected beam 

 goes at right angles to this along ad. By adjustments of the 

 mirrors b, c, d these two beams are made to pass round the 

 square to a again, where they are partly transmitted and partly 

 reflected, so that both now travel in the same direction along 



By carefully adjusting the mirror a these two beams are 

 superposed so as to produce interference-fringes, which are 

 observed through the telescope D. In actual experiment we 

 found that the two beams produced by the semitransparent 

 mirror were of unequal brilliancy, the reflected beam being the 

 more intense ; and in order to obtain the best effect we found 

 it necessary to move that mirror in the direction of the trans- 

 mitted beam so as to decrease the angle of incidence, thereby 



* Phil. Trans, vol. clxxxiv. (1893). 



7063S 



