12 Larmor, Physical Aspect of the Atomic Theory. 



pound vibrating system is nowhere treated by Young ; but, 

 at any rate, he would have been quite prepared to predict 

 the modern phenomena of anomalous optical dispersion 

 as arising from the sympathetic vibrations of the mole- 

 cules of the material medium. Later (1817), on again 

 taking up the subject of Physical Optics, in his Article in 

 the Encyclopcsdia Britannica on *' Chromatics," he seems 

 to have been dominated by the new puzzles connected 

 with his principle of polarization due to transverse vibra- 

 tion, and this mode of explaining dispersion is dropped. 



What are these parts of bodies — solid, liquid, or 

 gaseous — which are thus taken to be susceptible of per- 

 manent vibrations of their own? At any rate, the 

 hypothesis implies a thoroughly discrete structure of 

 matter. And it is perhaps remarkable that Young was 

 not tempted to ascribe the slowing of the period belonging 

 to a given wave length to the mere loading of the aether 

 by inert structureless particles of matter, in the manner 

 of the explanation of dispersion which Cauchy and Poisson 

 afterwards imposed on optics ; he went to the root of the 

 question, in the Newtonian manner (as he remarks in the 

 appended Corollary and Scholium), by ascribing to these 

 particles free periods of intrinsic vibration, and therefore 

 definite and identical structures. 



Nothing further is heard of this point of view until 

 the origin of the lines of the spectrum, and the mechanism 

 of the production of the dark Fraunhofer lines, were 

 discussed between Stokes and W. Thomson in 1854.* 

 In i860, in introducing the practical discovery of spectrum 

 analysis by Kirchhoff and Bunsen to the notice of the 

 British public, Stokes based his dynamical explanation of 

 selective absorption of radiation on the simple remark 

 that, when the waves passing through the medium are 



* G. G. Stokes, " Math, and Phys. Papers," vol. 4, 1904, appendix 



