1892.] a Discussion concerning Ether and Matter. 101 



medium, work done on a moving mirror, and the laws of reflexion 

 and refraction as modified "by motion, are considered. 



It is found that the law of reflexion is not really obeyed in a 

 relatively moving medium, though to an observer stationary with 

 respect to the mirror it appears to be obeyed, so far as the first order 

 of aberration magnitude is concerned ; but that there is a residual 

 discrepancy involving even powers of aberration magnitude, of an 

 amount possibly capable of being detected by very delicate observa- 

 tion. 



The following statements are made and justified : — 



(1.) The planes of incidence and reflexion are always the same. 



(2.) The angles of incidence and reflexion, measured between ray 

 and normal to surface, usually differ. 



(3.) If the mirror is stationary and medium moving, they differ by 

 a quantity depending on the square of aberration magnitude, 

 i.e., by 1 part in 100,000,000 ; and a stationary telescope, if 

 delicate enough, might show the effect. 



(4.) If the medium is moving and mirror stationary, the angles 

 differ by a quantity depending on the first power of aberra- 

 tion magnitude (1 part in 10,000), but a telescope moving 

 with the mirror will not be able to observe it; for the 

 commonplace aberration caused by motion of receiver will 

 obliterate the odd powers and leave only the even ores ; the 

 same as in case (3). 



(6.) As regards the angles which the incident and reflected waves 

 make with the surface, they differ in case (3) by a first 

 order magnitude, in case (4) by a second order magnitude. 



(6.) At grazing incidence the ordinary laws are accurately obeyed. 

 At normal incidence the error is a maximum. 



(7.) The ordinary laws are obeyed when the direction of drift is 

 either tangential or normal to the mirror ; and are disobeyed 

 most when the drift is at 45°. 



(8.) In general, the shape of the incident wave is not precisely pre- 

 served after reflexion in a moving medium. To a parallel 

 beam the mirror acts as if slightly tilted ; to a conical beam 

 as if slightly curved. But either effect, as observable in the 

 result, is almost hopelessly small. 

 (9.) Similar statements are true for refraction, assuming Fresnel's 

 law. 



The possibility of obtaining first order effects from general ethereal 

 motion by means of electrical observations is considered. 



