2 Scientific Proceedings, Royal Dublin Society. 



the downward directed ray which involved necessarily that the 

 section and its image in the reflector did not accurately overlie 

 one another. In rocks of fairly coarse grain this did not signify ; 

 but, in those of finer grain, an unpleasant overlapping of the 

 colours of adjacent crystals occurred in the plane of incidence 

 and reflection. In all the forms of the apparatus described, there 

 was also required a separate polarizer to polarize the beam entering 

 the illuminator. 



Recently it occurred to me to try the effect of a simple vertical 

 illuminator which I saw described in Messrs. Watson's Catalogue, 

 consisting of a single small cover-glass contained within a collar 

 to be inserted just above the object-glass. The cover-glass is to 

 be inclined so that rays entering an aperture in the front of the 

 collar are, in part, reflected by the cover-glass (which can be rotated 

 on a horizontal axis into the suitable inclination), and thence pass 

 downward through the object-glass and illuminate the opaque 

 object being examined. The rays finally reaching the eye (return- 

 ing through the object-glass much the way they came) are for 

 the most part transmitted through the transparent reflector. 



In this arrangement the illumination is evidently vertical ; and, 

 the parallax in the plane of incidence and reflection involved with 

 the use of prisms or opaque reflectors should be absent. I did 

 indeed suggest such a mode of getting out of the difficulty in my 

 former paper, but had not then given it trial. 



On applying one of these very simple and inexpensive illumi- 

 nators to the purpose described, I found that its use was in every 

 way satisfactory. The quantity of light transmitted is sufficient 

 even without the use of a lens to strengthen the beam. As an 

 artificial source, a mantle-burner is excellent, and shows colour 

 well. There is no appreciable parallax ; and even small microlithic 

 felspars in basalts may be seen, each glowing with its own colour 

 and with sharp margins. 



From the first I noted the added advantage that, with this mode 

 of illumination, the use of a polarizer is unnecessary. Even using 

 a horizontal beam and the glass at 45° to axis of microscope, the 

 polarization of the ray reflected from the uncovered speculum metal 

 is very complete, as may be seen by examination with a double- 

 image prism, or more simply by rotating the analyser. But when 

 the source of light is elevated above the horizontal level of the 



