1890.] upon Magnetic Changes of Length in Wires. 



Fig. 1. 



471 



The magnetising coil. CC, is supported by a stopper, A, which is inserted into the 

 bottom of the coil. Through an axial hole in A is screwed a brass rod, terminating 

 in a stirrup, S, beneath which is fixed a hook, H, for the suspension of weights. A 

 second brass rod, suspended by a pin at P, passes freely through a stopper, B. The 

 wire under experiment, X, is clamped between the ends of the brass rods. The 

 knife-edge at the bottom of the stirrup acts upon a brass lever, E, one edge of which 

 turns upon the knife-edge D, the other actuating a short arm, E, attached perpen- 

 dicularly to the back of the mirror M. The mirror turns upon knife-edges about 

 its horizontal diameter. By means of a lantern the image of a fine wire is, after 

 reflection from the mirror, projected upon a distant vertical scale and serves as an 

 index. Dimensions : — SD = 10 mm., DE = 170 mm., ME = 7 mm., distance from 

 mirror to scale = 4706 mm., each scale division = 0'64 mm., length of X = 100 mm. 



successively attached to it were equivalent to 1950, 1600, 1170, 819, 

 585, and 351 kilos, per square cm. of section. 



The nickel wire was 100 mm. long, and 0*65 mm. in diameter ; it 

 was supplied by Messrs. Johnson and Matthey. The loads under 

 • which it was examined were 2310, 1890, 1400, 980, 700, and 420 kilos, 

 per sq. cm. 



