THE PROOF OF THE EARTH'S MOTION 287 



observer on the earth is just the same as that of a man sailing 

 in a boat ; and if light, like the wind, has a measurable speed, 

 as Roemer and Newton supposed, then obviously the apparent 

 direction of the stars would be an effect compounded of the 

 double movement of the earth and the light. 



Up to that time observers had been going upon the theory 

 that the flash of light from the stars is instantaneous therefore, 

 that the motion of the earth need not be taken into considera- 

 tion. To them the problem in fixing the location of the star 

 was much the same as that of the professor of mechanics in 

 a familiar tale, through whose window one night a bullet came 

 crashing from across the green. By observing carefully the hole 

 in the window and in the wall opposite, and calculating the 

 path of the projectile, it was an easy matter for a mathematician 

 to determine from which student's room the careless shot had 

 come. 



But suppose that the observer is aboard a vessel flying along 

 at tremendous speed ; then, in calculating the point from which 

 the shot is fired, the speed of the vessel must be taken into 

 consideration. If its speed be great enough, the hole in the 

 wall opposite the window will lag a little behind the point where 

 the bullet would have struck had the vessel been standing still. 

 Obviously the lag will be greatest when the shot is fired directly 

 at right angles to the path of the ship's motion ; it will grow 

 less the nearer the path of the projectile agrees with the path 

 of the ship, and disappear entirely when the agreement is 

 complete. 



All this is precisely what Bradley observed. If the effect 

 he had detected is due to the finite speed of light, then it was 

 not difficult to calculate from his measurements that the velocity 

 of light must be very close to ten thousand times the speed of 

 the earth. The earth shoots along its track at nineteen miles 

 per second ; ten thousand times that is one hundred and ninety 

 thousand. The figure was considerably higher than Roemer's, 

 very close to Newton's. There could no longer be any doubt 

 that the speed of light is finite, and that its measure was known. 

 It is to be noted that Roemer had computed the velocity of 

 reflected solar light, Bradley the direct light of the stars. For 

 a time it was supposed that the discrepancy might be in this 

 difference ; but from the discussion of a thousand eclipses of 

 Jupiter's satellites, instead of the forty which Roemer could 



