312 POLARISATION. 



light what we gain in perfection of polarisation. Moreover, it is 

 only to be expected that the objective image must lose somewhat 

 in sharpness through the interposition of so large a body, and, of 

 course, to greater extent, the less perfectly its faces are polished. 

 On the other hand, as stated by H. von Mohl, 1 we shall obtain a 

 larger and more convenient field of view by this position than by 

 any other. 



Valentin 2 endeavours to explain the imperfect darkening of 

 the field of view by the divergence of the rays after their 

 passage through the objective. In the sentence we refer to, it is 

 stated that when this divergence exceeds the angle of aperture of 

 the Nicol, ordinary rays pass through together with the extra- 

 ordinary ones, and consequently give a bright field of view with 

 crossed Nicols. This is, of course, quite correct ; but since the 

 limiting angle of the total reflexion amounts to about 12, or, 

 according to Valentin, even 14 18 for the axis of the Nicol, it 

 would be entirely unjustifiable to assume so important a diver- 

 gence. The image-forming cones of light, which proceed from the 

 objective, have their base situated upon its second principal plane, 

 and occupy a linear aperture of 3 to 5 mm., according to the 

 magnification. Their maximum inclination is determined by the 

 size of the diaphragm in the eye-piece setting, and in the Micro- 

 scopes known to us does not exceed 4 ; adding half the aperture 

 of a single cone of light say 1 we find that the greatest 

 possible inclination of the marginal rays to the axis of the 

 Microscope is about 5. Hence there can be no question of the 

 transmission of the ordinary rays. 



The supposition of a too great divergence of the emerging cones- 

 of light would not, moreover, suffice to account for the phenome- 

 non in question, even for the case where it was perfectly well 

 founded. For since each cone of light is directed towards a definite 

 point in the image, the middle portion of the surface of the image, 

 as far as it is projected by rays of less than 12 inclination, 

 should appear completely dark with crossed Nicol s. A faint light 

 becoming more and more bright towards the edge would first 

 be noticeable at the margin, and this marginal portion might 

 easily be cut off by a corresponding diminution of the field of view. 



Tli subsequent inferences which Valentin draws from this 



1 Poggendorff's " Annalen," Bd. cviii. (1859), p. 181. 



2 Ib. p. 97. 



