an Electric Micrometer. 

 (2) Calibration. 



549 



If we assumed that the screw and the divisions of the disk 

 were accurate in all respects, we should still have to determine 

 the joint leverage of the three levers in order to measure any 

 movement ; but bv the following method of calibrating we 

 can obtain a knowledge of the value of the units on the disk 

 without making even these assumptions. 



In fig. 3 is given diagrainmatically the plan of the apparatus 

 involved. Fix to the end of the short arm of L 3 a glass plate 

 (a), and close to it a small lens (6) supported from the frame 

 of' the apparatus. The plate was a microscope glass slide 

 chosen for its good surface ; it was held fast by a screw and 

 nut between the lever and a brass washer. 



u; .■■ ■ 'i \ \ > -~l 



The lens was let into a saw-gate in a brass bar and fixed 

 there by shellac ; the brass bar was clamped by screws to a 

 strono- upright bar screwed to the wooden frame. Thus the 

 glass plate is fixed to the lever, and the lens is fixed to the 

 wooden frame, both are therefore shielded from the vibrations 

 of the table. 



Put up a Bunsen burner B, with sodium light and a glass 

 plate r to reflect the rays normally to the glass surface, then 

 an observer at the microscope M will, under favourable con- 

 ditions, see at the glass surface the centre of a system of 

 Newton's rings. The rings can be made to open out or con- 



