88 Dr. J. Kerr's Electro-optic 



a good many repetitions of this operation are sometimes re- 

 quired before the charge is sensibly free from specks of solid 

 matter. 



When the cell has to be charged with a new liquid, the 

 press is unscrewed, the planks and panes are removed, the 

 faces of the block are cleaned, two clean panes are laid against 

 the mouths of the tunnel, and held there while the cell is 

 thoroughly rinsed with a proper solvent. The panes 

 are removed, the faces of the block are again cleaned, two 

 clean panes are placed as before, and the rinsing is re- 

 peated. Half-a-dozen thorough rinsings with ether and alco- 

 hol are generally sufficient to prepare the cell perfectly for any 

 new liquid. 



3. Optical Compensator of Strained Glass. — In the follow- 

 ing experiments, I require continually to introduce definite and 

 very faint birefringent actions at some point between polarizer 

 and analyzer. The pieces which I employ for this purpose 

 are perfectly rectangular slips, all cut out of one carefully 

 selected plate of glass -^ of an inch thick, and all of the same 

 dimensions, width f of an inch, length 7 inches. When such 

 a slip is extended, it acts upon the transmitted light as a posi- 

 tive uniaxal with axis along the line of tension ; and when it 

 is compressed it acts as a negative uniaxal with axis along 

 the line of compression. These optical actions are in most 

 cases very definite and sensibly pure, though perhaps never 

 perfect as the actions of good natural crystals. 



4. Hand Compensator. — This is one of the slips (3), worked 

 by hand in the manner described in my former papers. It is 

 held between the first and second Nicols, immediately in front 

 of the latter, with its faces perpendicular to the ray, and its 

 length inclined at 45° to the plane of polarization. Kept 

 steadily in position, it is strained by an effort of hands or fin- 

 gers at each end, the axes of the couples applied being parallel 

 to the ray, so that one edge of the slip is extended and the 

 opposite edge compressed. According to the statement already 

 made in (3), the extended and compressed parts of the slip act 

 respectively as positive and negative uniaxals with axes parallel 

 to the edges. When the glass is chosen with ordinary care, 

 and preserved from any considerable changes of temperature 

 during the observation, these optical actions are (to sense) 

 perfectly pure, tension of one slip being always totally neutra- 

 lizable by parallel compression or perpendicular tension of 

 another slip. 



5. Fixed Compensator. — This is one of the standard slips 

 (3), which hangs freely in a constant vertical position from a 

 purposely constructed stand. To give the required effect of 



