Manchester Memoirs, Vol. xlvii. (1903), No. 7. 3 



it was found, if the " magic " effect was to be produced. 

 As the polishing proceeded, the mirror (which, by the 

 way, is a hand one, about yin. in diameter) was repeatedly 

 examined and found to possess the " magic " property, 

 and increasingly so as the polishing continued. 



It now occurred to me that this property might be the 

 result of either of two things, or both combined, vis., 

 (i) the varying resistance to flexure, due to the different 

 thicknesses of the metal (by reason of the raised design 

 at the back), causing the surface to assume an unequal 

 curvature, and thus exposing some portions of the surface 

 to the grinding and polishing actions more than others ; 

 or (2) that the metal itself, again by reason of the raised 

 design, might be more easily polished in some portions 

 than in others. With the object of determining to 

 which of these two causes the effect was due, an area of 

 about 9 square inches was reground and polished by 

 hand, using the least possible pressure. The result over 

 this area was an almost total absence of " magic" effect. 

 This result was not a little surprising, as the writer 

 had rather held that the effect was due, to a great extent 

 at least, to a difference of density, and of course it is 

 quite possible that this is a factor, but a very slight one, 

 in view of the results afterwards obtained. 



The mirror was now repolished in the machine for 

 several hours, using considerable pressure, with the result 

 that the " magic " effect was fairly good over the whole 

 surface, and this can only have resulted from the first 

 named cause, vie:, varying resistance to flexure, as the 

 previous polish by hand was about equal on the limited 

 surface to that on the main one. 



It now occurred to me, seeing that the effect was 

 undoubtedly due, for the most part at least, to the unequal 

 flexure during polishing, that straining the mirror might 



