166 Mb KELLAND, ON THE DISPERSION OF LIGHT, 



some inverse function of the distance, we have the first factor also very- 

 large, and have therefore reason to conclude that the velocity is a max- 

 imum. 



This consideration then to which we are driven by the observed 

 phenomena is perfectly consistent with itself, and it does not appear 

 to bear about it any a priori absurdity. That it will be found a matter 

 of some difficulty to adapt it to the explanation of certain phenomena 

 which are reduced to mathematical computation by the contrary sup- 

 position, is no argument against its validity. 



But let us examine how this result works when applied to the ex- 

 planation of other phenomena. It is clearly reasonable to conclude, that 

 since the action of a material substance on the particles of tether causes 

 a diminution of their density, that diminution will increase in propor- 

 tion as the number of material particles increases, and consequently for 

 the same kind of material particles the density of the tether bears some 

 inverse ratio to that of the substance. 



Suppose the particles of glass through which light is transmitted, 

 compressed by a force acting parallel to the axis of s : this, according 

 to the above conclusion, would lead us to infer that the particles of 

 tether were more dilated in the direction of ss, than in other directions. 



Now it is evident that whatever law of force we conceive as the 

 true one, it must diminish as the distance increases, and consequently 

 Fii-) must be negative. 



Also if we consider the vibrations transversal (an hypothesis which 

 will be noticed hereafter), we have the square of the velocity of trans- 

 mission along the axis of .r, of a vibration parallel to s, equal to 



12 \^{r) -f- ^^ l%^\ Ix- nearly, 

 whilst the analogous expression for a wave transmitted along the axis of 



is 



12{<^(;-) +^lx^]l^-, 



