Mercury Globules as Test Objects. By J. W. Gordon. 15 



from his ow,n cornea, but an image reflected from the under face of 

 this new globule, which I will, when further referring to it, speak 

 of as the speculum globule, to distinguish it from the object 

 globule on the stage. 



It may be convenient at this point to invite you to consider the 

 nature of the optical arrangement thus set up. It is, of course, a 

 very common observation that when two mirrors are placed on 

 opposite walls facing one another we get a great number of 

 successive reflections producing the appearance to the observer 

 placed between them of a long vista of mirror frames and many 

 repetitions of his own head. The same thing would, of course, 

 happen if our mirrors were convex mirrors. But in that case the 

 successive images would very rapidly diminish in size. In the case 

 of plain mirrors the successive images diminish in apparent size as 

 the result of perspective, but in the case of convex mirrors they 

 would diminish not only as the result of perspective but also by 

 reason of the magnifying power of the mirrors themselves. This 

 is what happens in the case of two mercury globules lacing one 

 another. The observer looking, as indicated in fig. 5, past the 



Fig. 5. 



speculum globule into the face of the object globule, sees there an 

 image of the inner face of the speculum globule and in that image, 

 which I will speak of as an image of the first order, he sees an 

 image of the second order of the object globule itself as reflected 

 in the face of the speculum globule. This second order image is 

 of necessity a very small image, for it has undergone reduction 

 in size, first by the speculum globule and then by the object globule 

 itself. If now we interpose a lens between these two globules 

 we do not prevent in any way the interchange of reflections 

 between them. The phenomena are, of course, somewhat compli- 

 cated by the magnifying power of the lens, but are not otherwise 

 affected by it. We are thus led to expect that if the optical 

 system of the Microscope were interposed between the two 

 globules of fig. 5, we should still have the second order image of 

 the object globule seen in its own surface. This is what actually 

 happens, and in one of the Microscopes upon the table this evening 

 you will find an arrangement of this sort set up and a brilliant 

 second order image of the object globule exhibited to view. 



