JAMES BRADLEY 125 



appeared that the displacements had the expected yearly 

 period, but quite unexpectedly did not correspond to a paral- 

 lax. For the motion was not opposed to that of the earth; 

 on the contrary, the star appeared displaced every time in the 

 same direction as the earth's motion. 



The following explanation was soon found by Bradley 

 for this unexpected displacement, which was then called 

 'aberration.' If, as Roemer had observed, the light coming 

 from a star takes time to be propagated in the telescope, and 

 the telescope is moving with the earth across the direction 

 of the light ray, it will be necessary to incline the telescope 

 a little in the direction of motion, in order to receive the ray 

 along the axis of the telescope, and the star will thus appear 

 displaced in the direction of motion to this extent. This is 

 the same phenomenon as when we move along under rain 

 falling vertically; we become wetter in front than behind, 

 as if the rain were coming from in front of us. If this ex- 

 planation is correct, the angle through which the star appears 

 to be diverted, must be given by the relationship between the 

 earth's velocity and the velocity of light, and this is found to 

 be the case. 



This agreement meant new insight of the greatest import- 

 ance. Roemer's measurement of the velocity of light was 

 brought into an entirely new kind of connection with Coper- 

 nicus' and Kepler's knowledge of the earth's motion; and 

 this amounted to a confirmation of all these facts. At the 

 same time, the discovery of aberration gave us a new pheno- 

 menon of light, which will be of importance for all time as 

 regards questions of the ether, and of absolute motion 

 through space. 



The angle of aberration is very small, in accordance with 

 the very great velocity of light; it amounts only to twenty 

 seconds of arc. The fijced-star parallaxes originally sought 

 for, are still smaller; even for the nearest stars they amount 

 to less than one second of arc. It was therefore not until 



