24 Stokes, on the Nature of the Rontgen Rays. 



ture which may be roughly compared to a flexible chain 

 with a small weight at the end of it. Suppose you have 

 vibrations communicated to such a chain at the top ; 

 they travel gradually to the bottom, and near the bottom 

 produce a disturbance which deviates more from a 

 simple harmonic undulation. So, if a vibration is com- 

 municated to what I will call the tail of the molecule of 

 uranium, it may give rise to a disturbance in the ether 

 which is not of a regular periodic character. I conceive, 

 then, that you have vibrations produced in the ether, 

 not of such a permanently regular character as would 

 constitute them vibrations of light, and yet not of so 

 simple a character as in the Rontgen rays — something 

 between. And accordingly there is enough irregularity 

 to allow the ethereal disturbance to pass through black 

 paper, and enough regularity on the other hand to make 

 possible a certain amount of refraction. You can also 

 obtain evidence of the polarisation, and, consequently, 

 of the transverse character of these rays. 



According to the theory of the nature of the Rontgen 

 rays which I have endeavoured very briefly to bring 

 before you, we have here, as I think, a system the 

 various parts of which fit into one another. You start 

 with the Rontgen rays, which consist, as I conceive, 

 of an enormous succession of independent pulses ; you 

 pass to the Becquerel rays, which are still irregular, but 

 are beginning to have a certain amount of regularity ; 

 and you end with the rays which constitute ordinary 

 light. According to this theory, the absence of diffrac- 

 tion in the Rontgen rays is explained, not by supposing 

 they are rays of light of excessively short wave length, 

 but by supposing they are due to an irregular repetition 

 of isolated and independent disturbances. So far as I 

 know, the view I have been led to form as to the nature 

 of refraction, and which forms an integral portion of the 



