the Refractive Index and Dispersion of Glass. 413 



will be found possible so to adjust the refractive index of 

 the liquid, that on moving the edge slowly across the 

 observing telescope the whole field, including the faces of 

 the prism and those of the cell to the right and to the l«ft 

 of the prism, darkens simultaneously and uniformly. Under 

 these circumstances the prism clearly no longer exerts any 

 prismatic action — in other words, an equality between the 

 indices has been attained, and a reading for the refractive 

 index of the liquid as it stands upon the Pulfrich prism may 

 be taken at once. 



Tt will be convenient now to describe the form of cell and 

 the immersion fluid used in these experiments. Before 

 doing so, however, it may be worth while to point out what 

 are the conditions which it is desirable should be satisfied. 

 In the first place it must be possible for light to pass into 

 the cell, to strike the upper surface of the base-plate at 

 grazing incidence and to be refracted down through the 

 base-plate and thence into the prism of the refractometer. 

 It will be evident from elementary geometrical consider- 

 ations, that if the width of the base-plate be equal to the 

 aperture of the Pulfrich prism upon which it stands, th^n, 

 owing to the finite thickness of the base-plate, a considerable 

 portion of the aperture of the Pulfrich prism will not be 

 filled with light unless the glass forming the base-plate be of 



Fia-. 2. 



9 



i 



high refractive index. A glance at fia. 2, giving a side 

 view of the cell in position upon the Pulfrich prism, will 

 explain the reason for this at once. 



Of the full aperture AB of the Pulfrich prism it will be 



