2 Transactions of the Society. 



globules of mercury mounted between two cover-glasses in Canada 

 balsam, and the only addition which I have made to this apparatus 

 — all of which, so far, has been already described— is that I have 

 fitted upon the supporting collar a pierced platform, which forms 

 a convenient support for an inspection lens when the stop has 

 to be adjusted, and for a camera when a photograph is to be 

 taken. Fig. 1 is a sketch of the complete apparatus, with an 

 inspection lens in place, and fig, 2 is another sketch, showing 

 my camera in position. 



Before proceeding actually to discuss my photographs, I may, 

 perhaps, say a word or two about the theory of this instrument as 

 I understand it. It is, of course, quite commonly understood that 

 in some way the image formed by a Microscope is dependent upon 

 the illumination of the object itself, and that when iine detail is in 

 question a better result may, as a rule, be secured by illuminating 

 with a wide than with a narrow cone of lioht. Furthermore, it is 

 perfectly obvious that when the stage of the IMicroscope is occupied 

 by an object of high or low refractive index, the nature of the 

 illumination is only in part dependent upon the angle of the 

 incident beam. An objective of wide angle may be filled with 

 light by the beam from a condenser of low angle if part of the 

 light has been sufficiently deflected from its original course by 

 reflection, refraction, or diffraction as it passes the stage of the 

 instrument. Hence, when we speak of the illumination of an 

 object in the Microscope, we mean the light received by the 

 objective, not the light received by the object, for a very large pro- 

 portion of the objects with which we are familiar illuminate them- 

 selves. Stained objects are almost the only exception. Even, 

 polished ol)jects are not exceptions to this rule, for a highly - 

 polished surface, to be visible at all, must be placed at an angle 

 to the plane of the stage, and then it deflects the incident light 

 in such a way as to be distinguishable from the field by the 

 different angle at which its light crosses the optical axis — that is 

 to say, it is in an optical sense self-luminous. While polished 

 objects thus become effectively self-luminous by reflection, trans- 

 parent objects become visible, when visible at all, by refraction of 

 the light which they transmit. This refraction makes them self- 

 luminous in the same sense, for a new ray, not to be found in the 

 light from the field, starts from such an object as its point of 

 origin. And even stained objects — among which 1 include such 

 as show by simple opacity — may become self-luminous if they 

 exhibit the phenomena of repetition, for any form of ruled surface 

 produces diffraction along certain axes, and the diffracted light so 

 originating behaves exactly like reflected or refracted light of 

 equal intensity in the formation of the image. Thus in determining 

 the illumination of an object, the behaviour of the objective is, a& 

 a rule, of greater importance than that of the condenser, for in most 



