Observations on various Liquids. 241 



a pulley. A very acute prism of flint-glass is laid flat upon 

 the face of the other flange, and is attached rigidly to the 

 bobbin by a screw cap, so that a ray of light, received along 

 the axis of the tubular shank, enters the prism normally as it 

 leaves the tube, and is then deviated through a small angle by 

 refraction. The bobbin has its shank supported in proper 

 bearings fixed at the edge of a whirling table, and can be made 

 to revolve round its geometrical axis very rapidly. When a 

 point is viewed through the bobbin by an eye placed close 

 behind the prism, the rotation of the bobbin transforms the 

 luminous point into a fine and faintly luminous circular line. 

 With my particular prism, this ring of light is rather too large 

 to be well commanded as a whole by the observer's eye ; but a 

 good view may always be had of as much as half the circum- 

 ference. There is of course a little dispersion ; but this is of 

 no consequence. 



55. Arrangements. — These are the same as in 51, with a 

 few additions. All the optical pieces are shown in the adjacent 

 diagram. 



a 3 



JM 



H K 





llE 



The flame L is turned broadside on to the cell; and the 

 horizontal slit in the screen P Q is reduced to a very small 

 width. The lens Gr is so placed as to give a real image of the 

 slit at the centre of the cell, the image being a narrow and bril- 

 liantbancl, lying along the axis of the electric field, and extending 

 from ball to ball. The second lens H assists the observer's vision. 

 S is the special eyepiece (54), carefully placed beforehand so 

 that the axis of the shank, which is also the axis of rotation, 

 coincides with the line of the ray L M N. The piece K, between 

 cell and analyzer, is a square plate of glass a quarter of an inch 

 thick, which is permanently strained in a small screw press, the 

 line of compression being parallel to the faces of the plate. 

 The use of K is the same as that of the hand compensator ; but 

 its action is more uniform, and can be made very much 

 stronger. 



As in 51, the first Nicol M is fixed with its principal 

 section at 45° to the horizon, and (the plate K being away) 

 the second Nicol is fixed at extinction. 



