.540 Theory of a Contractile JEther. 



at the polarizing-angle, and, in Jamin's notation, 



M = e sin cf>. 

 According to his observations for flint glass, 



€ = •017, = 59°-44. 

 Whence I = -000069. 



If we take Quincke's * observations, we find, from his table 

 for reflexion at the surface of flint glass, the value I = '000096. 

 Thus the value of I for glass is less than '0001, or one terth 

 the amount of light that is reflected from the black film of a 

 soap-bubble. Hence the light reflected from the bubble being 

 brighter ought to show more colour than the light reflected 

 from glass near the polarizing-angle. 



Two explanations may be given of the fact that colour has 

 not been noticed — except, possibly, in the rough observation 

 of my own already referred to — in either case. The one is 

 that, owing to its faintness, it has escaped the notice of 

 observers who were not specially looking for it ; the other 

 lies in the fact that there seems some reason to suppose that 

 our eyes are sensitive to light before they appreciate distinc- 

 tions of colour. Either of these would, 1 think, be sufficient 

 to account for the facts, and would allow us to believe that 

 the explanation of the Jamin-efFect given by Lorenz is the 

 true one, though possibly further experiments, which I hope 

 shortly to undertake, may be necessary to prove this. This 

 elliptic polarization depends greatly, it is true, as Wernicke t 

 has shown lately, on the nature of the surface and of the means 

 taken to polish it ; but he comes to the conclusion " that it is 

 a general property of bodies modified by the presence of a 

 surface-film, but not entirely explained by that." 



This surface-film of Wernicke's is, however, quite different 

 from the surface-layer of aether of variable density considered 

 by Lorenz. Wernicke's film is caused by the presence of 

 foreign matter; the layer of varying density is necessary to the 

 transition from air to glass, and it is only a question of how 

 thick this layer is in comparison with the wave-length. 



We conclude, then, that the theory here put forward 

 accounts satisfactorily for reflexion and refraction both by 

 transparent bodies and by metals, also for double refraction 

 and dispersion, including the anomalous dispersion of such") 

 substances as cyanin, fuchsin, or the other anilin dyes, 

 while it leads, in addition, to the correct expression for the 

 velocity of light in a moving medium. As already stated, I 

 hope to treat, in another paper, the consideration of the pro- 

 perties of quartz, the rotation produced by sugar, and the 

 rotation in a magnetic field. 



* " Optische Experimental Untersuchungen," Pogg. Ann. Band 127, 

 128, &c. t Wied. Ann. xxx. p. 468. 



