386 THE AMERICAN MONTHLY [Dec 



In the direction 1, 2, 3, appear the first three imnges. 

 No. 1 is from the glass surface, No. 2 is from the »ilver 

 surface, and No. 3 is from the silver and air surfaces. 



Move a card along A towards 1, and No. 3 disappears 

 tirst, No. 2 immediately after, and No. 1 when the card 

 reaches that point. So much for their origin. 



It will be asked, perhaps, how the images can alter 

 tlieir position when the mirror is revolved in the plane 

 of A. They cannot. The mirror A B has parallel sur- 

 faces. Microscope mirrors and most plate-glass mirrors 

 are not parallelised, but are, at the best, "optically" flat- 

 tened, and may be regarded as wedges. 



It is then easily seen how images approximate and 

 retire when the mirror is revolved. 



Let us give surfaces A and B an inclination of l°(Fig. 

 2). Then viewing a small object at E (close to the eye) 

 one image appears towards 1, i.e., at right angles to A, 

 and another in the direction E 2 — li° from E 1, which, 

 aft'^r being refracted to 1° in the glass, is reflected at 

 ri<;ht ano-les from surface B. 



There is another image nearer the letter A, but, as it 

 follows the same laws apparently as the others, save that 

 it is a real double reflection, we need not consider it. If 

 this mirror is revolved in the plane of A, of course No. 1 

 image will remain still. No. 2 and subsequent images 

 will revolve with the mirror round No. 1. If we exager- 

 ate this wedge shape of our mirror, we can see that at a 

 peculiar angle these images can be made to superimpose. 

 Let the signs be as before (Fig. 3) and the images whose 

 rays pass respectively from to 1 and 2' will be reflect- 

 ed to E as one image. I should imagine the third image 

 to arrive at E through 1. but I have not yet worked this 

 out. Of course, placing the eye at and the object at 

 E would be equivalent to revolving the mirror. The im- 

 ages vary slightly in size owing to their various distances. 



No. 2 is the brightest except at great obliquity. 



