is polarized at the Surface of the Metals. 299 



Brewster. In fact, he desciibed the tints of the reflected images 

 as succeeding each otlier by simple alternations from the greatest 

 to tlie smallest reirangibilitv, whereas I tlierein ascertained evi- 

 dently the whole series of the rings reflected and transmitted. 

 He indicated these tints as polarized, one in the plane of refrac- 

 tion, and the other in the perpendicular direction ; whereas I 

 found them polarized at equal distances from this plane, the one 

 in the direction of the primitive polarization, and the other oa 

 the opposite side, confornial)ly to the theory of oscillations: hence 

 it followed also that a single reflection upon silver could not im- 

 press on the natural light any determinate polarization. Having 

 spoken to I\I. Arago of this apparent contradiction, he assured 

 me that he had observed that the light reflected bv silver, like 

 the other metallic bodies, always experiences a very sensible 

 partial polarization, according to the plane of incidence ; and he 

 ^ave nie ' a piece of polished silver on which in fact I could as- 

 certain this property. This seemed to agree with the indica- 

 tions of Mr. ijrewster, and to be contrary to mine. But as facts 

 can never be really contrary to facts, 1 inquired what could be 

 the difference in the elements of the two observations ; and I at 

 length thought that the different nature of the polish might have 

 an influence on the mode of polarization exercised by the me- 

 tallic laminae. This was in short perfectly confirmed by ex- 

 perience. 



We may give a polish to a metal in two ways, viz. by the ham- 

 mer or by friction. The first mode consists in beating the me- 

 tallic plate on a polished anvil v.'ith a polished hammer ; after 

 which a lustre is given to its surface bv rubbing it with glove- 

 leather impregnated with a very fine polishing red. This pro- 

 cess when applied to silver gives it a very fine whiteness; but the 

 images are always a little undulated, and as if frothy at the edges. 

 Ill the abundant reflection of light which takes place, we do 

 not recognise the lively and brilliant polish of looking-glasses. 



The polish by friction is that which is given to telescope- 

 mirrors : they are first rubbed on a stone of a very soft grain 

 (hone-stone), and a brilliancy is given to their suriace by rubbing 

 them on pitch coated with putty of tin. Then, if the operation 

 has been well conducted, the images are well defined, lively, and 

 the reflection has all the specular appearance. 



Now, 'by a very remarkable circumstance those two kinds of 

 polish do not act in the same way upon incident light. I do 

 not speak here of the (juantity more or less considerable of light 

 which the surfaces reflect, but of the mode itself in which thev 

 act upon the luminous molecules, and of the direction in which 

 they polarize them. When the surface of sdver or of any other 

 metal has received the specular polish, it produces by regular 



reflection 



