Dec, 26, 1878] 



NATURE 



179 



2. For the sake of precision we will take a special 

 example. Suppose a luminous source with such an 

 adjustment as to emit a parallel beam of light,^ and let 

 the luminous source be supposed put in motion in a 

 direction at right angles to the path of the beam. Then 

 on the basis of the corpuscular theory (according to which 

 light consists of projected particles) since the particles or 

 corpuscles, according to the laws of motion, partake 

 necessarily of the motion of the body emitting them ; so, 

 therefore, if the corpuscular theory were true, the path of 

 the emitted beam of light would be exactly the same in 

 direction as if the luminous body were at rest. So, there- 

 fore, if we imagine a screen of the same breadth as the 

 luminous beam to be placed at a distance, so that this 

 screen is exactly illuminated when the luminous source is 

 at rest, then (according to the corpuscular theory) this 

 screen would also be exactly illuminated when both it 

 and the luminous source were put in motion with equal 

 velocities (in parallel paths) in the same direction. 



3. This, however, will not be the case if the undulatory 

 theory be true, for it is a known fact that waves emitted 

 in a medium do not partake of the motion of the body 

 emitting them. For when once a wave has left the body, 

 the wave is dependent solely on the medium for its propa- 

 gation and is not influenced by the motion of the body 

 one way or the other. It follows, therefore, that in the 

 case of a moving luminous source emitting waves trans- 

 versely to its path, the waves forming the parallel beam 

 will be left behind, or will not partake of the motion of the 

 luminous source. The waves will form a slanting track 

 of light which will no longer strike exactly the opposed 

 distant screen, but will fall somewhat to the rear of it. 

 The luminous beam which, when the screen and source 

 were at rest, was exactly eclipsed or intercepted by the 

 screen, will (when the screen and source are in motion) 

 commence to escape behind the edge of the screen, or 

 the eclipse will no longer be total. 



4. Here, therefore, we should have in principle a simple 

 and decisive test between the two theories, provided iin- 

 superable practical difficulties do not stand in the way of 

 carrying it out (for which object probably various methods 

 would suggest themselves). 



In order to contrast further the ditferen*: effects that 

 would be produced in the case of the two theories (cor- 

 puscular and undulatory), we may consider various 

 possible cases of relative motion, also the effect when the 

 beam is received directly in a telescope, or in the eye. 

 We have ah-eady considered the case where the beam is 

 observed objectively (by the use of a screen), which we 

 may call Case I. 



5. Case II. — We may now consider the case when a 

 telescope 2 is used. We will take the above example of 

 a luminous source in motion, emitting a parallel beam of 

 light at right angles to its path, and we will imagine that 

 this beam is received in a distant stationary telescope, 

 placed normal to the path of the moving luminous source, 

 so that the beam flashes down the axis of the telescope 

 at the instant of the passage of the luminous source. 

 Then we have to compare the effects produced in the 

 case of the undulatory and corpuscular theories. On the 

 undulatory theory, waves emitted by a luminous source do 

 not partake of the motion of the source, so that at the 

 instant when the wave of the beam (singling out a parti- 

 cular wave) flashes down the axis of the telescope at the 

 moment of passage of the luminous source, the source 

 will have akeady moved on a distance from the point 

 where the wave left it, this distance representing that 

 traversed by it during the time the light took to pass from 

 the source to the telescope ; and the source is therefore 

 seen out of its true position by precisely that amount. 



necZr,^°.°*'k" a j}ara//r/ beam of Ught for sImpHcity, though it is net 

 necessary to the principle. «- j> & 



alon^tc*^^* '''I '^'^" produced by aberration is the same with the eye 

 Ka?e%^=.W» ,t'*'!r^°P^'^"'-'*^P''=*'"'° consider the lattei, as its larger 

 scale enables the effect to be visualised better 



This is perfectly evident, and the correction for this error 

 in the estimate of position (of the value above indicated) 

 constitutes the well-known "equation of light." 



6. We have now to consider what takes place on the 

 corpuscular theory. Here the projected corpuscles will 

 partake of the motion of the source. Singling out, there- 

 fore, one of the corpuscles that flashes down the axis of 

 the telescope at the instant of the passage of the luminous 

 source ; this corpuscle will possess the transverse 

 velocity of the source that emitted it, and therefore the 

 corpuscle will not pass straight down the axis of the 

 stationary telescope, but in its passage will deviate late^- 

 rally from that axis. The telescope would accordingly 

 have to be inclined in order that the corpuscle might 

 pass along the axis. This deviation of the corpuscle 

 from the axis of the telescope will cause the luminous 

 source to be viewed out of its true position, and it is 

 easily seen that this visual error in the estimate of the 

 position of the source on the corpuscular theory is pre- 

 cisely the same in amount as the previous error (due to 

 a different cause) on the undulatory theorj'. Indeed, the 

 error on the corpuscular theory is simply a case of 

 "aberration" due to the relative motion of the telescope 

 and light), and the correction for it, according to known 

 principles, is the same as the other correction on the 

 undulatory theor}', termed " equation of light."' It is a 

 remarkable fact, therefore, that though the path of the 

 light in its transit is very different in the case of the two- 

 theories, the visual error in the estimate of the position 

 of the object is the same, so that this error cannot itself 

 serve as a test between the two theories. There is, how- 

 ever, one marked distinction between the two theories ; 

 for while on the undulatory theory a position of the 

 telescope normal to the path of the moving luminous 

 source, causes the flash of the beam to pass down the 

 axis of the telescope ; on the corpuscular theory, on the 

 other hand, the telescope has to be inclined in order 

 that the flash of the beam may pass down its axis. Here, 

 therefore, we have a definite physical effect serving as a 

 point of distinction between the two theories. 



7. Case III. — We will now take the case when both 

 the luminous source and the distant observer are iiv 

 motion, moving with equal velocities in parallel paths 

 alongside each other in the same direction. Here on 

 the corpuscular theory, in which case the corpuscles par- 

 take of the motion of the source, since the whole system 

 therefore moves with equal velocity, the whole systera 

 will therefore be relatively at rest ; so that the light will 

 pass across and enter the telescope just as it would have 

 done if everything were at rest, or there will be no pecu- 

 liarity in the passage of the light whatever on the corpus- 

 cular theory. It is widely different on the undulatory 

 theory, for here the beam of light passing between the 

 source and tffe telescope will be left behind in the medium, 

 and therefore in the first place, the moving telescope, in 

 order to catch the parallel beam, will have to be placed 

 back a certain distance in the rear ; for since the light 

 takes a slanting track between the source and telescope 

 (the degree of slant depending on their common velocity), 

 the telescope to intercept the light, can no longer be 

 placed exactly opposite the luminous source. The dis- 

 tance the telescope wll require to be placed back evidently 

 must be equal to that traversed by it during the time the 

 light takes to pass from the source to the telescope. 

 Secondly, the light on the undulatory theory will suffer 

 aberration in passing along the tube of the telescope, 

 owing to the latter being in motion relatively to the light; 

 no such aberration taking place on the corpuscular theory, 

 since the corpuscles are moving at the same velocity as 

 the telescope. Thirdly, on the undulatory theory, there 

 will be a correction necessarj-, due to the motion of the 

 luminous source ("equation'of light"); such correctioa 

 not being required on the corpuscular theor)', since on 

 that theory the light emitted partakes of the motion of 



