400 Prof. Stokes on the Metallic Reflexion 



indicated by the character of the rings, while it is very evident 

 for red light, becomes much more considerable in the highly 

 refrangible colours. Now specular iron is almost opake for light 

 of all colours, but as it gives a red streak it appears that the sub- 

 stance-colour is red ; and, in fact, it is known that very thin 

 laminae are blood-red by transmitted light. Accordingly, the 

 chromatic variation of the change of phase corresponds to that 

 of the intense absorbing power. 



The light reflected by specular iron is not extinguished by 

 analysation, whatever be the angle of incidence; but at the 

 angle of incidence which gives the nearest approach to complete 

 polarization, a quantity of blue light is observed to remain. This 

 has been explained by comparing specular iron to a substance of 

 high dispersive power, so that the polarizing angle for red light 

 is considerably less than for blue ; and accordingly on increasing 

 the angle of incidence, the light (which is here supposed to be 

 analysed so as to retain only the portion polarized perpendicularly 

 to the plane of incidence), while it becomes much less copious 

 near the polarizing angle, becomes at the same time of a decided 

 blue colour*. I believe, however, that the blue light is mainly 

 due to the chromatic variation of the metallicity, the medium, 

 considered optically, being much more metallic for blue light 

 than for red, though it may in some measure be due to the cause 

 previously assigned. 



Specular iron is a good example of a substance forming a 

 connecting link between the true metals and substances like 

 safflower-red. It resembles metals in the circumstance that the 

 absorbing power, as inferred from the chromatic variation of the 

 metallicity, and as indicated by the tint of the streak, is not 

 subject to the same extensive chromatic variations as in the case 

 of colouring matters like safflower-red. It resembles safflower- 

 red in being sufficiently transparent with respect to a portion of 

 the spectrum to allow the connexion between the metallicity and 

 the substance-colour to be observed; whereas the substance- 

 colour of metals is not known from direct observation, except, 

 perhaps, in the case of gold, which in the state of gold-leaf lets 

 through a greenish light. 



I am now able to bring forward a striking confirmation of the 

 relation which seems to exist between the light reflected as if 

 from a metal, and that absorbed with great energy. On reading 

 M. Haidinger's paper, of which the title has been already quoted, 

 I was particularly interested by finding crystallized permanga- 

 nate of potash mentioned among the substances which exhibit 

 distinct surface- and substance-colours. I had previously noticed 



* See Dr. Lloyd's Lectures on the Wave-Theory of Light, part 2. p. 18. 



