Prof. Kelland on the explanation of Dispersion. 377 



stated his conclusions very differently in other memoirs. For 

 instance, in the Comptes Renclus for 1836, p. 343, he states 

 thatFresnel is justified in saying that the vibrations of aethe- 

 real molecules generally lie "in the planes of the waves ; and 

 to the same effect in the Comptes Rendns for 1S38, p, 866. 

 The fact is, that the whole depends on the law of force, as I 

 have shown more than once. If the law be that of the in- 

 verse square of the distance, A is negative, and the vibrations 

 are transversal only. (See my memoir on Dispersion, p. 180.) 



The second result of Mr. O'Brien presents nothing which 

 belongs to our present purpose. The third is, tliat the di- 

 spersion of light may he completely accounted for, 'without having 

 recourse to the hypothesis offnite intervals. This result is re- 

 ferred to the hypothesis, that the material particles influence 

 the vibrations of the eethereal ones by their attractions or re- 

 pulsions. 



I think I have shown clearly that the principle of the ex- 

 planation has not been lost sight of by the advocates of the 

 interval theory. The point at issue then is simply this. Will 

 the direct attractions or repulsions of the particles of matter, 

 symmetrically immersed in those of aether, affect the velocity 

 of vibration of the latter? According to my views they will 

 not; and consequently in my second hypothesis I confine my at- 

 tention to their indirect effect. But that effect depends on their 

 mutual distances, and \\\\x^ fnite intervals, not indeed of the 

 particles of aether, but of those of matter, necessarily play a 

 conspicuous part. Until, then, it shall have been proved 

 " that the hypothesis of finite intervals cannot be correct," 

 which I do not think can well be done by means of such as- 

 sumptions as those on which the equations, adopted by Mr. 

 O'Brien, depend (see p. 205, or Cauchy's Ex. vol. iii. p. 190), 

 the real chfference between the received theory and that be- 

 fore us is this; that the former rejects the direct attraction of 

 the particles of matter as producing no effect on the time of 

 vibration of a particle of aether, whilst it retains the variation 

 of action due to the distance between the particles, either of 

 one system or of the other ; the latter retains the Ibrmer and 

 rejects the latter. Now as it has been shown (on the hypothesis 

 of the Newtonian law) that the term which we reject fs zero, it 

 is incumbent on those who refuse to receive our conclusions 

 to siiow that what we retain is either zero or imaginary. 

 Until this, or something cfjuivalent, shall have been done, the 

 foriner sujjporters of the theory of finite intervals will hardly 

 be inclined to give it up. 



I have the honour to be, dear Sir, 



Your obliged Servant, 



P. Kelland, 



