X Proceedings. [Z^rav///;^;- i6tli, igo2. 



however, uniformly convex, the portions opposite the rehef on 

 the baciv being approximately plane. Writing with reference to 

 the peculiar property of these mirrors. Person says: — "The rays 

 "reflected from the convex portion diverge, and give but a 

 "feebly illuminated image, while, on the contrary, the rays 

 " reflected from the plane portions of the mirror preserve their 

 "parallelism, and appear on the screen as an image by reason 

 " of their contrast with the feebler illumination of the rest of 

 "the disc." Professors Ayrton and Perry give the following 

 explanation : — " A preliminary operation in polishing the surface 

 " consists of scoring the cast disc in every direction with a sharp 

 " tool. The thicker portions with relief ornament offer more 

 " resistance to the pressure of the tool than the thin flat portions 

 " which tend to yield and form at first a concave surface, but 

 " this, by the reaction of its elasticity, rises afterwards and forms 

 "a slightly convex surface, while the more rigid thick portions 

 " are comparatively little affected." This explanation was 

 accepted by Professor Sylvanus Thompson at a recent lecture on 

 the subject. 



Mr. Southern thought that, if this explanation were correct, 

 it would be difficult to see how a very small area of thin metal 

 lying in the middle of a thick area could be reproduced in the 

 reflection, since it would be strongly supported by the thick 

 metal and would not yield appreciably under the pressure of the 

 scoring tool ; moreover, the rim, which is of the same thickness 

 as the thicker portions of the design, would be distinctly shown 

 in the reflection, which is not the case ; again, in some large 

 specimens the portions of the surface opposite broad areas of 

 relief on the back appear to be very slightly concave, for which 

 the above explanation would not account. 



It seems probable, therefore, that there must be some other 

 or additional explanation, and Mr. Southern suggested that the 

 peculiar properties of the mirrors might be caused by the varying 

 densities of the thick and thin portions of the casting, due to 

 unequal cooling ; the thicker portions, having cooled more 

 slowly, would be softer and more readily abraded in the process 



