1G4 UNDULATORY THEORY OF LIGHT. 



rays upon a wliite surface, when the same points will become manifest by their 

 difference of illuminating power. The first method is best, especially if the eye 

 be assisted by a lens, but the second is that which was used by the earliest 

 observers. 



We cannot detect the interferences of light by observing periodical maxima 

 and minima, like the beats in music, because of the almost inconceivable short- 

 ness of the undulations. But if the waves of light Avere as long as the waves 

 of sound, interferences might easily be made to manifest themselves, something 

 in the manner of the scintillation of the stars, though Avith a regularity which 

 that phenomenon does not possess. 



B'-fort' proceeding now to a more particular inquiry into the laws of the inter- 

 fi-rence of luminous waves, it is proper to make two or three preliminary ex- 

 planations. TIk^ phenomena compel us to the assumption that the molecular 

 movements in these waves are normal to the direction of progress ; that is, to 

 the direction of vision. In other words, they are in the plane of the wave itself, 

 and at right angles to the ray. If we suppose that all ethereal tremors have 

 this character, we must account for the fact by presuming that the ether is 

 nearly incompressible. If this is the case, the vibrations of the luminous body, 

 at its surface, may move latcraUy the whole stratum of ethereal particles which 

 is most nearly in contact with it, though they produce little if any motion per- 

 pendicular to the surface. 



If MN, for example, be the surface of the luminous body, A, A, A, &;c., the 

 row of ethereal particles next it. A', A', A', &c., the row beyond this, and so on, 



the aiTangement of these particles will, on the 

 ^"-^ A'f JE' vC" ^' -0' ^^ principle of etiuilibrium, be such that the dis- 



-Q-' ^1^^' ^-^' d^" ^o^p p tances of adjacent particles shall be equal. If 

 ^-^p-^^^o o o d the molecular vibrations of the surface JMN are 



^—ir^^'^. °^ <^/i 9, 9i incapable of driving the particles A, A, A, &c., 

 j^j ]ff du-ectly outward toward the plane or A',A',A', 



Fio-. 36. &c., on account of the very difficult compressi- 



bility of the ether, they may, nevertheless, 

 move them all sideways in the direction AB. Let the entire force of this 

 movement b<^ represented by AB. Join AA', and di-aw BG perpendicular to 

 it The force AB may be resolved into the two forces CB and AC, of which 

 the second is directed toward the centre of A'. This again may be resolved 

 into the two, AD and DC, of which the first is normal to MN, and, by hy- 

 pothesis, produces no sensible movement. But DC is parallel to MN, and, as 

 all the other particles in the stratum A',A', and are simultaneously acted upon 

 by similar forces, they will all move in the direction of DC, without changing 

 their distances from each other. 



It Is not necessary to suppose that the ether has absolutely no compressibility. 

 In fact, if it had none at all, it could have no elasticity ; or, what is, singularly 

 inough, practically the same thing, its elasticity would be infinite. But its 

 eora[)ressibility must be esteemed very slight, and its elasticity accordingly very 

 liigh, not merely because of the n<^cessity of admitting lateral or transverse 

 A-ibrations, but because of the immense veloiity of light. It is easy to see that, 

 if the ether were totally incompressible, the velocity of light (if in such circum- 

 stances there could be any such thing as light) would be infinite ; that is to say, 

 any movement in the ether, if it could be produced at all, must be produced 

 simultaneously tlu-ough the whole extent of the ether. In proportion as com 

 pression is easy, the rapidity of the propagation of a disturbance (density re- 

 maining the same) must be less. The immense velocity of light affords, therefor , 

 a strong ground for believing that the compressibility of the ether is very small. 

 Still, it is hardly conceivable that there should exist absolutely no molecular 

 movements normal to the wave at all ; and, in fact, the existence of such vibra- 



