INVERTEBEATA, CIIYPTOGAMIA, MICKOSCOPY, ETC. 



875 



an object have parallel lines upon it, and these lie in a direction from 

 the back to the front of the stage, the diffraction spectra will bo 

 arranged in a row at right angles to the lines, and half of the spectra 

 which would otherwise have been admitted will be cut off by the 



Fig. 88. 



Fig. 89. 



z 



o 



2 



o 



Object. 



Spectra. 



prism (Figs. 88 and 89). If, however, the object is turned round so 

 that the lines run from left to right, the row of spectra will bo 

 admitted (Figs. 90 and 91), and there will consequently bo no 

 diminution of aperture. 



Fig. 90. 



Fig. 91. 



Object. 



Spectra. 



If the number of lines in the object were doubled, the distance 

 between the spectra would also be doubled, and the spectrum No. 1 

 in Fig. 89 would occupy the place of No. 2, and would thus fall 

 outside the field, and there would bo no resolution. On turning tho 

 object round, however (see Figs. 90 and 91), the spectrum No. 1, 

 though more widely separated from the central illuminating beam 

 by reason of the greater fineness of the lines would still fall within tho 

 field (occupying the jdace of No. 2) and tho object would be resolved. 



It is therefore of practical importance to see that the object is 

 properly placed when a binocular is used. 



Apertures exceeding ISC in Air. — Many microscopists still 

 experience a difficulty in gras])ing tho idea of an object-glass having 

 an upcrture " exceeding 180' in air," but a little consideration should 

 di8j)cl any difficulty. 



Fig. 92 represents the theoretical maximum of 180" in oil or other 

 homogeneous fluid. It is a semicircle, enclosing 30 spaces, radiant 

 from the point A (of 6° each), S])read out as a fan, through which the 

 diffraction spectra emanating from A may bo su])posed to pass. 



If wo now substitute for tho oil a fluid, such as icater, having a 



