882 SUMMARY OF CUERENT RESEARCHES RELATING TO 



will tire you more than a whole day's work in a dark room by lamp- 

 Hght." 



Aperture Puzzles. — Another puzzle turns on the statement some- 

 times made that it is not necessary to have an objective of 1-0 N.A. 

 (180° air) to resolve 96,000 lines to the inch as shown by the Aperture 

 Table; that it can be effected by a dry objective of say 0-50 N.A. 

 (60° air). 



The way in which this feat is supposed to be accomplished is by 

 attaching a truncated cone A to the cover-glass as shown in fig. 213, 

 the connection being made by balsam, oil, &c. Here the first diffrac- 



FiG. 213. Fig. 214. 



C-gl.l 



A.p. 



tion spectrum and the dioptric beam which leave the object (Amphi- 

 pleura pellucida) at an angle of 82° in glass emerge from the cone at 

 an angle of only say 60° in air, and are therefore collected by a dry 

 objective of approximately that aperture. 



The explanation simply is that by connecting the cone with the 

 cover-glass we have an immersion objective, with the difference that 

 (1) instead of using a hemispherical piece of glass to cause a pencil 

 of 82° in the glass to emerge as one of moderate air angle, to be 

 taken up by a dry objective above it, a conical one has been substi- 

 tuted, and that (2) in place of attaching the extra piece of glass once 

 for all to the rest of the optical combination, as in an ordinary im- 

 mersion objective, it has to be attached to each slide examined, at a 

 great sacrifice of convenience ! The cone has, moreover, the disad- 

 vantage, as compared with the hemisphere, that whilst the latter will 

 readily collect to a focus the whole pencil of 82°, and thus allows of 

 the real delineation of all kinds of objects, the cone will only collect 

 two very narrow partial pencils of equal and opposite obliquities and 

 will not bring these to a proper focus. The problem is in fact only an- 

 other mode of stating the old mare's-nest of the " hemisphere puzzle," 

 disguised by the substitution of a truncated cone for the hemisphere. 



The same effect, though with more dispersion, can be obtained 

 by refraction (instead of reflection) through a truncated prism of 

 isosceles section and suitable inclination of the refracting faces 

 (fig. 214), and whatever form is employed the only essential conditions 

 are that two infinitely narrow beams (the incident beam and the first 

 diffraction pencil) shall have equal and opposite inclinations u to the 

 axis within the front medium, and that they should be deflected, by 



