( 657 ) 



XLII. On the Conversion of Relief by Inverted Vision. .ZtySir DAVID BREWSTER, 

 K.H., D.C.L., F.R.S., and V.P.R.S. Edin. 



(Read 6th May 1844.) 



UNDER the name Conversion of Relief , an expression first used by Mr WHEAT- 

 STONE, I include all those optical illusions which take place in the vision of cameos 

 and intaglios, of elevations and depressions, whether they are produced with 

 opaque or transparent bodies, on surfaces with or without shadows, in reflected 

 or transmitted light, while using one or both eyes, or by erect or inverted 

 vision. In these various forms of the phenomenon, the illusion is modified by 

 certain secondary causes, which were regarded both by Mr WHEATSTONE* and 

 myself f as primary causes ; so that we were led away, each in a different direc- 

 tion, from the right path of inquiry. 



The phenomenon occurs in its most general and simple form, when it 

 is produced by viewing a shadowless depression, or elevation, made in an 

 extended surface, through an inverting microscope, or the inverting eye-piece 

 of a telescope, and at an angle intermediate between and 90. In so far as I 

 know, the phenomenon has never been thus limited, and, consequently, no ex- 

 planation of it has ever been given. That which I shall now submit to the So- 

 ciety is capable of the most rigorous demonstration ; and when it is once in our 

 possession, we can have no difficulty in recognising the secondary causes which 

 increase or diminish the influence of the primary one, and which, in its absence, 

 are sometimes the immediate cause of the illusion. 



Let A, Fig. 1, be a deep spherical concavity, and A', Fig. 2, a high spherical 



Fig. 1. Kig. 2. 





convexity in an extended horizontal table MN, M'N', and let them be shadowless^ 

 or illuminated by a quaquaversus light, like that of the sky. If the observer, placed 

 at a moderate distance, view these objects in the directions E A, E' A', either with 



* Phil. Trans., 1838, pp. 383, 384. 



t Edin. Trans., Vol. xv. p. 365 ; Edin. Journal of Science, Vol. iv. p. 97 ; and Lettert on 

 Natural Mayic, p. 98. 



VOL. XV. PART IV. 8 P 



