MEMORANDA. 165 



objective, even with 180° of divergence. In balsam there 

 must be the same dispersion of light; but beyond a certain 

 obliquity, it cannot emerge. The refractive index of Canada 

 balsam is 1 540, and that of such flint glass as I use for 

 covers 1'550 ; with these Mr. Sollittwill find that the limit is 

 40° 30' from the perpendicular, and that below this all the 

 rays must be reflected back and be absorbed or escape at the 

 edges of the slide. How much is thus lost cannot be assigned 

 without knowing how the intensity of this dispersed light 

 varies at different obliquities ; but if it be uniform, then the 

 balsam unquestionably cuts off a portion of it whose quantity 

 is measured by the difference between a hemisphere and a 

 segment whose amplitude is 81°, or • 76 of the whole. In 

 air the light will pass almost to 90°. 



The first argument by which Mr. SoUitt endeavours to 

 overthrow this inference is, that he finds his measures of 

 aperture the same, whether he interposes in the course of the 

 light a plain slide or one with balsam. This must be the 

 case ; for every one knows that a ray incident on any system 

 of media, bounded by parallel surfaces, emerges parallel to its 

 original direction.* But though it will pass the media, even 

 up to 90°, its course in t/iemis always within the angle of total 

 reflection, and the light whose loss I have described comes not 

 by direct transmission, but by dispersion. How that disper- 

 sion is produced I do not now inquire ; but its existence is a 

 fact. 



His second argument is, that test objects are as well seen in 

 balsam as when dry. This is not in accordance with my 

 experience, and I believe is not generally admitted. It is 

 true that a body of irregular surface and high refractive power 

 is (especially as to its internal structure) best seen in balsam ; 

 but this is a special case, and does not apply to one which, 

 like the valve of a Pleurosiyma, is thin and flat. Indeed, 

 besides the point in question, this medium may be expected 

 to injure vision by weakening the dispersion and reflection of 

 the light. The refractive index is nearly that of quartz for the 

 ordinary ray, which is probably the same as that of silica, as it 

 exists in these valves, having no active or polarized light ;t if 

 it were exactly the same, neither the valve nor its markings 

 could be seen at all, and as it is, its action makes the latter 



* It is shifted a little sideways, which is the reason that we can use 

 these large apertures with covers of a certain thickness, but this can 

 make no change in the measure. 



t In this respect thej' contrast strongly with undoubted vegetable pro- 

 ductions, such as the siliceous integuments of grasses and Equiseta, the 

 hairs of JDeutzia, &c. 



