Conversion of Cameos into Intaglios, fyc. 105 



A in the same manner, and with nearly the same intensity as 

 C does, the elevation will fall down into a depression. The 

 cause of this is obvious : the application of the inverting eye- 

 piece produces no effect whatever, for both the sides of the 

 cavity are symmetrically illuminated. In moving round the 

 second candle C from its position C, so as to stand beside C, 

 it is curious to observe the progress of the deception by which 

 the depression is again changed into an elevation 



If, when the depression A, Fig. 17, is converted into an 

 elevation, we introduce a small unpolished opaque body M, 

 and place it either beside the hollow or in it, so that the body 

 M, and its shadow m, may be distinctly seen by the micro- 

 scope, we shall have the appearance shown in Fig. 18, the 

 elevation having sunk into a depression. This correction of 

 the deception arises from the introduction of a new illusion, 

 namely, that which arises from the shadow m; for it is evi- 

 dent, that, as the body M appears to project its shadow in the 

 direction M m, the luminous body must be supposed to be on 

 the same side D ; and the evidence that this is the case, is more 

 powerful than our knowledge that the candle is actually at C, 

 because it co-exists along with our perception of the depres- 

 sion A, whereas our knowledge of the situation of the candle 

 is an act of recollection. 



This correction of the delusion may be effected in another 

 manner, which is perhaps more complete. If, in place of the 

 unpolished body, we use a pin with a highly polished head, 

 as shown at M, Fig. 19, and then apply the inverting eye- 

 piece, we shall have the effect shown in Fig. 20, the cavity A 

 appearing depressed. The image s of the candle C being 

 seen by reflection in the polished head of the pin M, is seen 

 by the application of the eye-piece at s, on the right hand 

 side of M in Fig. 20, so that we immediately conceive, in 

 opposition to our previous knowledge, that the candle must 

 be at D ; and hence the elevation falls into a depression the 

 moment the pin head is pushed up into the field of view. The 

 shadow M m has also its influence in the present case. 



The next case in which this illusion is dispelled, is, when 

 the sense of touch corrects the deduction formed through the 

 medium of sight. Let the cavity A be raised into an eleva- 



